diff --git "a/logs/training_reasoning_v2.log" "b/logs/training_reasoning_v2.log"
new file mode 100644--- /dev/null
+++ "b/logs/training_reasoning_v2.log"
@@ -0,0 +1,47 @@
+🚀 BEAST MODE TRAINING
+Device: mps
+
+Loading Oculus from checkpoints/oculus_detection_v2/final...
+  ✓ Loaded projector from checkpoints/oculus_detection_v2/final
+[Oculus] Loading language model...
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+
Loading weights:   0%|          | 0/472 [00:00<?, ?it/s]
Loading weights:   0%|          | 1/472 [00:00<00:00, 4675.92it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.bias]
Loading weights:   0%|          | 1/472 [00:00<00:00, 1324.38it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.bias]
Loading weights:   0%|          | 2/472 [00:00<00:00, 1205.61it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.weight]
Loading weights:   0%|          | 2/472 [00:00<00:00, 1176.52it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.weight]
Loading weights:   1%|          | 3/472 [00:00<00:00, 1282.14it/s, Materializing param=text_decoder.bert.embeddings.position_embeddings.weight]
Loading weights:   1%|          | 3/472 [00:00<00:00, 1219.04it/s, Materializing param=text_decoder.bert.embeddings.position_embeddings.weight]
Loading weights:   1%|          | 4/472 [00:00<00:00, 1182.83it/s, Materializing param=text_decoder.bert.embeddings.word_embeddings.weight]    
Loading weights:   1%|          | 4/472 [00:00<00:00, 1169.96it/s, Materializing param=text_decoder.bert.embeddings.word_embeddings.weight]
Loading weights:   1%|          | 5/472 [00:00<00:00, 1234.27it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:   1%|          | 5/472 [00:00<00:00, 1223.47it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:   1%|▏         | 6/472 [00:00<00:00, 1210.42it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:   1%|▏         | 6/472 [00:00<00:00, 1107.07it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:   1%|▏         | 7/472 [00:00<00:00, 1213.58it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.bias]      
Loading weights:   1%|▏         | 7/472 [00:00<00:00, 1112.21it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.bias]
Loading weights:   2%|▏         | 8/472 [00:00<00:00, 1231.54it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.weight]
Loading weights:   2%|▏         | 8/472 [00:00<00:00, 1160.61it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.weight]
Loading weights:   2%|▏         | 9/472 [00:00<00:00, 1168.26it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.bias]      
Loading weights:   2%|▏         | 9/472 [00:00<00:00, 1157.30it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.bias]
Loading weights:   2%|▏         | 10/472 [00:00<00:00, 1126.41it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.weight]
Loading weights:   2%|▏         | 10/472 [00:00<00:00, 1093.75it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.weight]
Loading weights:   2%|▏         | 11/472 [00:00<00:00, 1110.33it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.bias]
Loading weights:   2%|▏         | 11/472 [00:00<00:00, 1101.87it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.bias]
Loading weights:   3%|▎         | 12/472 [00:00<00:00, 1096.31it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.weight]
Loading weights:   3%|▎         | 12/472 [00:00<00:00, 1092.60it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.weight]
Loading weights:   3%|▎         | 13/472 [00:00<00:00, 1117.13it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.bias]  
Loading weights:   3%|▎         | 13/472 [00:00<00:00, 1109.22it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.bias]
Loading weights:   3%|▎         | 14/472 [00:00<00:00, 1182.42it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.weight]
Loading weights:   3%|▎         | 14/472 [00:00<00:00, 1093.83it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.weight]
Loading weights:   3%|▎         | 15/472 [00:00<00:00, 1077.65it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:   3%|▎         | 15/472 [00:00<00:00, 1071.74it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:   3%|▎         | 16/472 [00:00<00:00, 1073.62it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:   3%|▎         | 16/472 [00:00<00:00, 1054.29it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:   4%|▎         | 17/472 [00:00<00:00, 1108.79it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.bias]      
Loading weights:   4%|▎         | 17/472 [00:00<00:00, 1072.36it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.bias]
Loading weights:   4%|▍         | 18/472 [00:00<00:00, 1106.48it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:   4%|▍         | 18/472 [00:00<00:00, 1102.47it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:   4%|▍         | 19/472 [00:00<00:00, 1136.02it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.bias]      
Loading weights:   4%|▍         | 19/472 [00:00<00:00, 1116.21it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.bias]
Loading weights:   4%|▍         | 20/472 [00:00<00:00, 1161.99it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.weight]
Loading weights:   4%|▍         | 20/472 [00:00<00:00, 1134.42it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.weight]
Loading weights:   4%|▍         | 21/472 [00:00<00:00, 1153.53it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.bias]
Loading weights:   4%|▍         | 21/472 [00:00<00:00, 1128.80it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.bias]
Loading weights:   5%|▍         | 22/472 [00:00<00:00, 1149.37it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.weight]
Loading weights:   5%|▍         | 22/472 [00:00<00:00, 1128.44it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.weight]
Loading weights:   5%|▍         | 23/472 [00:00<00:00, 1142.02it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.bias]  
Loading weights:   5%|▍         | 23/472 [00:00<00:00, 1139.86it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.bias]
Loading weights:   5%|▌         | 24/472 [00:00<00:00, 1182.51it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.weight]
Loading weights:   5%|▌         | 24/472 [00:00<00:00, 1164.77it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.weight]
Loading weights:   5%|▌         | 25/472 [00:00<00:00, 1205.41it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.bias]         
Loading weights:   5%|▌         | 25/472 [00:00<00:00, 1202.67it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.bias]
Loading weights:   6%|▌         | 26/472 [00:00<00:00, 1230.43it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.weight]
Loading weights:   6%|▌         | 26/472 [00:00<00:00, 1228.62it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.weight]
Loading weights:   6%|▌         | 27/472 [00:00<00:00, 1241.33it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.bias]    
Loading weights:   6%|▌         | 27/472 [00:00<00:00, 1223.00it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.bias]
Loading weights:   6%|▌         | 28/472 [00:00<00:00, 1227.26it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.weight]
Loading weights:   6%|▌         | 28/472 [00:00<00:00, 1223.62it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.weight]
Loading weights:   6%|▌         | 29/472 [00:00<00:00, 1222.50it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.bias]      
Loading weights:   6%|▌         | 29/472 [00:00<00:00, 1218.64it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.bias]
Loading weights:   6%|▋         | 30/472 [00:00<00:00, 1226.09it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.weight]
Loading weights:   6%|▋         | 30/472 [00:00<00:00, 1210.56it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.weight]
Loading weights:   7%|▋         | 31/472 [00:00<00:00, 1231.73it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:   7%|▋         | 31/472 [00:00<00:00, 1218.85it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:   7%|▋         | 32/472 [00:00<00:00, 1238.67it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:   7%|▋         | 32/472 [00:00<00:00, 1237.06it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:   7%|▋         | 33/472 [00:00<00:00, 1270.65it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.bias]      
Loading weights:   7%|▋         | 33/472 [00:00<00:00, 1267.73it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.bias]
Loading weights:   7%|▋         | 34/472 [00:00<00:00, 1294.41it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.weight]
Loading weights:   7%|▋         | 34/472 [00:00<00:00, 1259.17it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.weight]
Loading weights:   7%|▋         | 35/472 [00:00<00:00, 1277.00it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.bias]      
Loading weights:   7%|▋         | 35/472 [00:00<00:00, 1274.59it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.bias]
Loading weights:   8%|▊         | 36/472 [00:00<00:00, 1286.41it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.weight]
Loading weights:   8%|▊         | 36/472 [00:00<00:00, 1272.85it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.weight]
Loading weights:   8%|▊         | 37/472 [00:00<00:00, 1303.78it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.bias]
Loading weights:   8%|▊         | 37/472 [00:00<00:00, 1302.27it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.bias]
Loading weights:   8%|▊         | 38/472 [00:00<00:00, 1284.96it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.weight]
Loading weights:   8%|▊         | 38/472 [00:00<00:00, 1273.12it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.weight]
Loading weights:   8%|▊         | 39/472 [00:00<00:00, 1293.79it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.bias]  
Loading weights:   8%|▊         | 39/472 [00:00<00:00, 1279.57it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.bias]
Loading weights:   8%|▊         | 40/472 [00:00<00:00, 1279.39it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.weight]
Loading weights:   8%|▊         | 40/472 [00:00<00:00, 1264.91it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.weight]
Loading weights:   9%|▊         | 41/472 [00:00<00:00, 1279.38it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:   9%|▊         | 41/472 [00:00<00:00, 1266.38it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:   9%|▉         | 42/472 [00:00<00:00, 1280.68it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:   9%|▉         | 42/472 [00:00<00:00, 1255.91it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:   9%|▉         | 43/472 [00:00<00:00, 1272.90it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.bias]      
Loading weights:   9%|▉         | 43/472 [00:00<00:00, 1254.49it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.bias]
Loading weights:   9%|▉         | 44/472 [00:00<00:00, 1276.28it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:   9%|▉         | 44/472 [00:00<00:00, 1272.89it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:  10%|▉         | 45/472 [00:00<00:00, 1285.82it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.bias]      
Loading weights:  10%|▉         | 45/472 [00:00<00:00, 1284.58it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.bias]
Loading weights:  10%|▉         | 46/472 [00:00<00:00, 1307.50it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.weight]
Loading weights:  10%|▉         | 46/472 [00:00<00:00, 1294.25it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.weight]
Loading weights:  10%|▉         | 47/472 [00:00<00:00, 1287.29it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.bias]
Loading weights:  10%|▉         | 47/472 [00:00<00:00, 1285.10it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.bias]
Loading weights:  10%|█         | 48/472 [00:00<00:00, 1307.98it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.weight]
Loading weights:  10%|█         | 48/472 [00:00<00:00, 1301.13it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.weight]
Loading weights:  10%|█         | 49/472 [00:00<00:00, 1314.84it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.bias]  
Loading weights:  10%|█         | 49/472 [00:00<00:00, 1312.90it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.bias]
Loading weights:  11%|█         | 50/472 [00:00<00:00, 1303.54it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.weight]
Loading weights:  11%|█         | 50/472 [00:00<00:00, 1294.04it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.weight]
Loading weights:  11%|█         | 51/472 [00:00<00:00, 1306.82it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.bias]         
Loading weights:  11%|█         | 51/472 [00:00<00:00, 1300.27it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.bias]
Loading weights:  11%|█         | 52/472 [00:00<00:00, 1310.29it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.weight]
Loading weights:  11%|█         | 52/472 [00:00<00:00, 1308.47it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.weight]
Loading weights:  11%|█         | 53/472 [00:00<00:00, 1308.88it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.bias]    
Loading weights:  11%|█         | 53/472 [00:00<00:00, 1298.42it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.bias]
Loading weights:  11%|█▏        | 54/472 [00:00<00:00, 1307.57it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.weight]
Loading weights:  11%|█▏        | 54/472 [00:00<00:00, 1305.27it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.weight]
Loading weights:  12%|█▏        | 55/472 [00:00<00:00, 1299.93it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.bias]      
Loading weights:  12%|█▏        | 55/472 [00:00<00:00, 1288.70it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.bias]
Loading weights:  12%|█▏        | 56/472 [00:00<00:00, 1307.87it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.weight]
Loading weights:  12%|█▏        | 56/472 [00:00<00:00, 1304.51it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.weight]
Loading weights:  12%|█▏        | 57/472 [00:00<00:00, 1316.47it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:  12%|█▏        | 57/472 [00:00<00:00, 1307.04it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:  12%|█▏        | 58/472 [00:00<00:00, 1327.05it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:  12%|█▏        | 58/472 [00:00<00:00, 1310.11it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:  12%|█▎        | 59/472 [00:00<00:00, 1303.26it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.bias]      
Loading weights:  12%|█▎        | 59/472 [00:00<00:00, 1301.85it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.bias]
Loading weights:  13%|█▎        | 60/472 [00:00<00:00, 1311.51it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.weight]
Loading weights:  13%|█▎        | 60/472 [00:00<00:00, 1309.59it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.weight]
Loading weights:  13%|█▎        | 61/472 [00:00<00:00, 1310.99it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.bias]      
Loading weights:  13%|█▎        | 61/472 [00:00<00:00, 1308.76it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.bias]
Loading weights:  13%|█▎        | 62/472 [00:00<00:00, 1311.65it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.weight]
Loading weights:  13%|█▎        | 62/472 [00:00<00:00, 1279.46it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.weight]
Loading weights:  13%|█▎        | 63/472 [00:00<00:00, 1268.96it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.bias]
Loading weights:  13%|█▎        | 63/472 [00:00<00:00, 1267.05it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.bias]
Loading weights:  14%|█▎        | 64/472 [00:00<00:00, 1257.89it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.weight]
Loading weights:  14%|█▎        | 64/472 [00:00<00:00, 1248.54it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.weight]
Loading weights:  14%|█▍        | 65/472 [00:00<00:00, 1264.59it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.bias]  
Loading weights:  14%|█▍        | 65/472 [00:00<00:00, 1256.45it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.bias]
Loading weights:  14%|█▍        | 66/472 [00:00<00:00, 1253.28it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.weight]
Loading weights:  14%|█▍        | 66/472 [00:00<00:00, 1249.03it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.weight]
Loading weights:  14%|█▍        | 67/472 [00:00<00:00, 1252.36it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:  14%|█▍        | 67/472 [00:00<00:00, 1250.45it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:  14%|█▍        | 68/472 [00:00<00:00, 1262.53it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:  14%|█▍        | 68/472 [00:00<00:00, 1239.20it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:  15%|█▍        | 69/472 [00:00<00:00, 1244.88it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.bias]      
Loading weights:  15%|█▍        | 69/472 [00:00<00:00, 1240.23it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.bias]
Loading weights:  15%|█▍        | 70/472 [00:00<00:00, 1241.75it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:  15%|█▍        | 70/472 [00:00<00:00, 1236.08it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:  15%|█▌        | 71/472 [00:00<00:00, 1244.83it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.bias]      
Loading weights:  15%|█▌        | 71/472 [00:00<00:00, 1244.11it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.bias]
Loading weights:  15%|█▌        | 72/472 [00:00<00:00, 1260.59it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.weight]
Loading weights:  15%|█▌        | 72/472 [00:00<00:00, 1260.04it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.weight]
Loading weights:  15%|█▌        | 73/472 [00:00<00:00, 1276.57it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.bias]
Loading weights:  15%|█▌        | 73/472 [00:00<00:00, 1276.03it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.bias]
Loading weights:  16%|█▌        | 74/472 [00:00<00:00, 1292.51it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.weight]
Loading weights:  16%|█▌        | 74/472 [00:00<00:00, 1291.97it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.weight]
Loading weights:  16%|█▌        | 75/472 [00:00<00:00, 1308.40it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.bias]  
Loading weights:  16%|█▌        | 75/472 [00:00<00:00, 1307.85it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.bias]
Loading weights:  16%|█▌        | 76/472 [00:00<00:00, 1324.27it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  16%|█▌        | 76/472 [00:00<00:00, 1323.70it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  16%|█▋        | 77/472 [00:00<00:00, 1340.11it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.bias]         
Loading weights:  16%|█▋        | 77/472 [00:00<00:00, 1339.60it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.bias]
Loading weights:  17%|█▋        | 78/472 [00:00<00:00, 1356.00it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.weight]
Loading weights:  17%|█▋        | 78/472 [00:00<00:00, 1355.46it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.weight]
Loading weights:  17%|█▋        | 79/472 [00:00<00:00, 1371.67it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.bias]    
Loading weights:  17%|█▋        | 79/472 [00:00<00:00, 1371.12it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.bias]
Loading weights:  17%|█▋        | 80/472 [00:00<00:00, 1387.30it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 80/472 [00:00<00:00, 1386.75it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 81/472 [00:00<00:00, 1403.18it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.bias]      
Loading weights:  17%|█▋        | 81/472 [00:00<00:00, 1402.62it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.bias]
Loading weights:  17%|█▋        | 82/472 [00:00<00:00, 1419.03it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.weight]
Loading weights:  17%|█▋        | 82/472 [00:00<00:00, 1418.49it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.weight]
Loading weights:  18%|█▊        | 83/472 [00:00<00:00, 1434.87it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  18%|█▊        | 83/472 [00:00<00:00, 1434.30it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  18%|█▊        | 84/472 [00:00<00:00, 1450.40it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  18%|█▊        | 84/472 [00:00<00:00, 1449.83it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  18%|█▊        | 85/472 [00:00<00:00, 1466.03it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.bias]      
Loading weights:  18%|█▊        | 85/472 [00:00<00:00, 1465.47it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.bias]
Loading weights:  18%|█▊        | 86/472 [00:00<00:00, 1481.69it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.weight]
Loading weights:  18%|█▊        | 86/472 [00:00<00:00, 1481.08it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.weight]
Loading weights:  18%|█▊        | 87/472 [00:00<00:00, 1497.30it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.bias]      
Loading weights:  18%|█▊        | 87/472 [00:00<00:00, 1496.64it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.bias]
Loading weights:  19%|█▊        | 88/472 [00:00<00:00, 1512.84it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.weight]
Loading weights:  19%|█▊        | 88/472 [00:00<00:00, 1512.23it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.weight]
Loading weights:  19%|█▉        | 89/472 [00:00<00:00, 1528.40it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.bias]
Loading weights:  19%|█▉        | 89/472 [00:00<00:00, 1527.80it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.bias]
Loading weights:  19%|█▉        | 90/472 [00:00<00:00, 1543.92it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.weight]
Loading weights:  19%|█▉        | 90/472 [00:00<00:00, 1543.32it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.weight]
Loading weights:  19%|█▉        | 91/472 [00:00<00:00, 1559.39it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.bias]  
Loading weights:  19%|█▉        | 91/472 [00:00<00:00, 1558.79it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.bias]
Loading weights:  19%|█▉        | 92/472 [00:00<00:00, 1574.64it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.weight]
Loading weights:  19%|█▉        | 92/472 [00:00<00:00, 1574.02it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.weight]
Loading weights:  20%|█▉        | 93/472 [00:00<00:00, 1590.07it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  20%|█▉        | 93/472 [00:00<00:00, 1589.42it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  20%|█▉        | 94/472 [00:00<00:00, 1605.27it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  20%|█▉        | 94/472 [00:00<00:00, 1604.59it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  20%|██        | 95/472 [00:00<00:00, 1620.58it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.bias]      
Loading weights:  20%|██        | 95/472 [00:00<00:00, 1619.90it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.bias]
Loading weights:  20%|██        | 96/472 [00:00<00:00, 1635.88it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  20%|██        | 96/472 [00:00<00:00, 1635.16it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  21%|██        | 97/472 [00:00<00:00, 1650.87it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.bias]      
Loading weights:  21%|██        | 97/472 [00:00<00:00, 1650.16it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.bias]
Loading weights:  21%|██        | 98/472 [00:00<00:00, 1666.05it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  21%|██        | 98/472 [00:00<00:00, 1665.34it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  21%|██        | 99/472 [00:00<00:00, 1681.19it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  21%|██        | 99/472 [00:00<00:00, 1680.50it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  21%|██        | 100/472 [00:00<00:00, 1696.33it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  21%|██        | 100/472 [00:00<00:00, 1695.63it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  21%|██▏       | 101/472 [00:00<00:00, 1711.46it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.bias]  
Loading weights:  21%|██▏       | 101/472 [00:00<00:00, 1710.73it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.bias]
Loading weights:  22%|██▏       | 102/472 [00:00<00:00, 1726.33it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  22%|██▏       | 102/472 [00:00<00:00, 1725.63it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  22%|██▏       | 103/472 [00:00<00:00, 1741.40it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.bias]         
Loading weights:  22%|██▏       | 103/472 [00:00<00:00, 1740.69it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.bias]
Loading weights:  22%|██▏       | 104/472 [00:00<00:00, 1756.44it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.weight]
Loading weights:  22%|██▏       | 104/472 [00:00<00:00, 1755.75it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.weight]
Loading weights:  22%|██▏       | 105/472 [00:00<00:00, 1771.23it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.bias]    
Loading weights:  22%|██▏       | 105/472 [00:00<00:00, 1770.54it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.bias]
Loading weights:  22%|██▏       | 106/472 [00:00<00:00, 1786.14it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  22%|██▏       | 106/472 [00:00<00:00, 1785.41it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  23%|██▎       | 107/472 [00:00<00:00, 1801.08it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.bias]      
Loading weights:  23%|██▎       | 107/472 [00:00<00:00, 1800.35it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.bias]
Loading weights:  23%|██▎       | 108/472 [00:00<00:00, 1816.08it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.weight]
Loading weights:  23%|██▎       | 108/472 [00:00<00:00, 1815.34it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.weight]
Loading weights:  23%|██▎       | 109/472 [00:00<00:00, 1830.99it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  23%|██▎       | 109/472 [00:00<00:00, 1830.25it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  23%|██▎       | 110/472 [00:00<00:00, 1845.52it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  23%|██▎       | 110/472 [00:00<00:00, 1844.80it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  24%|██▎       | 111/472 [00:00<00:00, 1860.26it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.bias]      
Loading weights:  24%|██▎       | 111/472 [00:00<00:00, 1859.55it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.bias]
Loading weights:  24%|██▎       | 112/472 [00:00<00:00, 1875.07it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.weight]
Loading weights:  24%|██▎       | 112/472 [00:00<00:00, 1874.32it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.weight]
Loading weights:  24%|██▍       | 113/472 [00:00<00:00, 1889.73it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.bias]      
Loading weights:  24%|██▍       | 113/472 [00:00<00:00, 1888.94it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.bias]
Loading weights:  24%|██▍       | 114/472 [00:00<00:00, 1904.44it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.weight]
Loading weights:  24%|██▍       | 114/472 [00:00<00:00, 1903.71it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.weight]
Loading weights:  24%|██▍       | 115/472 [00:00<00:00, 1919.23it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.bias]
Loading weights:  24%|██▍       | 115/472 [00:00<00:00, 1918.49it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.bias]
Loading weights:  25%|██▍       | 116/472 [00:00<00:00, 1933.92it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.weight]
Loading weights:  25%|██▍       | 116/472 [00:00<00:00, 1933.21it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.weight]
Loading weights:  25%|██▍       | 117/472 [00:00<00:00, 1948.58it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.bias]  
Loading weights:  25%|██▍       | 117/472 [00:00<00:00, 1947.83it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.bias]
Loading weights:  25%|██▌       | 118/472 [00:00<00:00, 1963.10it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.weight]
Loading weights:  25%|██▌       | 118/472 [00:00<00:00, 1962.32it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.weight]
Loading weights:  25%|██▌       | 119/472 [00:00<00:00, 1977.61it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  25%|██▌       | 119/472 [00:00<00:00, 1976.85it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  25%|██▌       | 120/472 [00:00<00:00, 1991.84it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  25%|██▌       | 120/472 [00:00<00:00, 1991.05it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  26%|██▌       | 121/472 [00:00<00:00, 2006.17it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.bias]      
Loading weights:  26%|██▌       | 121/472 [00:00<00:00, 2005.31it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.bias]
Loading weights:  26%|██▌       | 122/472 [00:00<00:00, 2020.51it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  26%|██▌       | 122/472 [00:00<00:00, 2019.68it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  26%|██▌       | 123/472 [00:00<00:00, 2034.81it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.bias]      
Loading weights:  26%|██▌       | 123/472 [00:00<00:00, 2034.01it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.bias]
Loading weights:  26%|██▋       | 124/472 [00:00<00:00, 2049.12it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  26%|██▋       | 124/472 [00:00<00:00, 2048.23it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  26%|██▋       | 125/472 [00:00<00:00, 2063.35it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  26%|██▋       | 125/472 [00:00<00:00, 2062.57it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  27%|██▋       | 126/472 [00:00<00:00, 2077.78it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  27%|██▋       | 126/472 [00:00<00:00, 2076.98it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  27%|██▋       | 127/472 [00:00<00:00, 2092.09it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.bias]  
Loading weights:  27%|██▋       | 127/472 [00:00<00:00, 2091.30it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.bias]
Loading weights:  27%|██▋       | 128/472 [00:00<00:00, 2106.37it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  27%|██▋       | 128/472 [00:00<00:00, 2105.58it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  27%|██▋       | 129/472 [00:00<00:00, 2120.67it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.bias]         
Loading weights:  27%|██▋       | 129/472 [00:00<00:00, 2119.86it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.bias]
Loading weights:  28%|██▊       | 130/472 [00:00<00:00, 2134.97it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.weight]
Loading weights:  28%|██▊       | 130/472 [00:00<00:00, 2133.91it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.weight]
Loading weights:  28%|██▊       | 131/472 [00:00<00:00, 2148.98it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.bias]    
Loading weights:  28%|██▊       | 131/472 [00:00<00:00, 2148.15it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.bias]
Loading weights:  28%|██▊       | 132/472 [00:00<00:00, 2163.19it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  28%|██▊       | 132/472 [00:00<00:00, 2162.38it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  28%|██▊       | 133/472 [00:00<00:00, 2177.37it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.bias]      
Loading weights:  28%|██▊       | 133/472 [00:00<00:00, 2176.55it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.bias]
Loading weights:  28%|██▊       | 134/472 [00:00<00:00, 2191.48it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.weight]
Loading weights:  28%|██▊       | 134/472 [00:00<00:00, 2190.66it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.weight]
Loading weights:  29%|██▊       | 135/472 [00:00<00:00, 2205.60it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  29%|██▊       | 135/472 [00:00<00:00, 2204.73it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  29%|██▉       | 136/472 [00:00<00:00, 2219.39it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  29%|██▉       | 136/472 [00:00<00:00, 2218.54it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  29%|██▉       | 137/472 [00:00<00:00, 2233.38it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.bias]      
Loading weights:  29%|██▉       | 137/472 [00:00<00:00, 2232.51it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.bias]
Loading weights:  29%|██▉       | 138/472 [00:00<00:00, 2247.41it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.weight]
Loading weights:  29%|██▉       | 138/472 [00:00<00:00, 2246.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.weight]
Loading weights:  29%|██▉       | 139/472 [00:00<00:00, 2261.34it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.bias]      
Loading weights:  29%|██▉       | 139/472 [00:00<00:00, 2260.46it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.bias]
Loading weights:  30%|██▉       | 140/472 [00:00<00:00, 2275.20it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.weight]
Loading weights:  30%|██▉       | 140/472 [00:00<00:00, 2274.35it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.weight]
Loading weights:  30%|██▉       | 141/472 [00:00<00:00, 2289.00it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.bias]
Loading weights:  30%|██▉       | 141/472 [00:00<00:00, 2288.08it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.bias]
Loading weights:  30%|███       | 142/472 [00:00<00:00, 2302.77it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.weight]
Loading weights:  30%|███       | 142/472 [00:00<00:00, 2301.91it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.weight]
Loading weights:  30%|███       | 143/472 [00:00<00:00, 2316.69it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.bias]  
Loading weights:  30%|███       | 143/472 [00:00<00:00, 2315.64it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.bias]
Loading weights:  31%|███       | 144/472 [00:00<00:00, 2330.37it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.weight]
Loading weights:  31%|███       | 144/472 [00:00<00:00, 2329.46it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.weight]
Loading weights:  31%|███       | 145/472 [00:00<00:00, 2344.19it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  31%|███       | 145/472 [00:00<00:00, 2343.29it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  31%|███       | 146/472 [00:00<00:00, 2357.54it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  31%|███       | 146/472 [00:00<00:00, 2356.60it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  31%|███       | 147/472 [00:00<00:00, 2371.16it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.bias]      
Loading weights:  31%|███       | 147/472 [00:00<00:00, 2370.28it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.bias]
Loading weights:  31%|███▏      | 148/472 [00:00<00:00, 2384.76it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  31%|███▏      | 148/472 [00:00<00:00, 2383.80it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  32%|███▏      | 149/472 [00:00<00:00, 2398.40it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.bias]      
Loading weights:  32%|███▏      | 149/472 [00:00<00:00, 2397.43it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.bias]
Loading weights:  32%|███▏      | 150/472 [00:00<00:00, 2411.73it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  32%|███▏      | 150/472 [00:00<00:00, 2410.81it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  32%|███▏      | 151/472 [00:00<00:00, 2425.32it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  32%|███▏      | 151/472 [00:00<00:00, 2424.39it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  32%|███▏      | 152/472 [00:00<00:00, 2438.72it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  32%|███▏      | 152/472 [00:00<00:00, 2437.77it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  32%|███▏      | 153/472 [00:00<00:00, 2452.08it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.bias]  
Loading weights:  32%|███▏      | 153/472 [00:00<00:00, 2451.09it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.bias]
Loading weights:  33%|███▎      | 154/472 [00:00<00:00, 2465.27it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  33%|███▎      | 154/472 [00:00<00:00, 2464.28it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  33%|███▎      | 155/472 [00:00<00:00, 2478.73it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.bias]         
Loading weights:  33%|███▎      | 155/472 [00:00<00:00, 2477.83it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.bias]
Loading weights:  33%|███▎      | 156/472 [00:00<00:00, 2491.93it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.weight]
Loading weights:  33%|███▎      | 156/472 [00:00<00:00, 2490.94it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.weight]
Loading weights:  33%|███▎      | 157/472 [00:00<00:00, 2505.27it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.bias]    
Loading weights:  33%|███▎      | 157/472 [00:00<00:00, 2504.31it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.bias]
Loading weights:  33%|███▎      | 158/472 [00:00<00:00, 2518.66it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  33%|███▎      | 158/472 [00:00<00:00, 2517.69it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  34%|███▎      | 159/472 [00:00<00:00, 2532.09it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.bias]      
Loading weights:  34%|███▎      | 159/472 [00:00<00:00, 2531.17it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.bias]
Loading weights:  34%|███▍      | 160/472 [00:00<00:00, 2545.42it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.weight]
Loading weights:  34%|███▍      | 160/472 [00:00<00:00, 2544.50it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.weight]
Loading weights:  34%|███▍      | 161/472 [00:00<00:00, 2558.89it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  34%|███▍      | 161/472 [00:00<00:00, 2557.95it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  34%|███▍      | 162/472 [00:00<00:00, 2571.99it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  34%|███▍      | 162/472 [00:00<00:00, 2570.86it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  35%|███▍      | 163/472 [00:00<00:00, 2584.93it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.bias]      
Loading weights:  35%|███▍      | 163/472 [00:00<00:00, 2583.90it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.bias]
Loading weights:  35%|███▍      | 164/472 [00:00<00:00, 2598.14it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.weight]
Loading weights:  35%|███▍      | 164/472 [00:00<00:00, 2597.11it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.weight]
Loading weights:  35%|███▍      | 165/472 [00:00<00:00, 2611.38it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.bias]      
Loading weights:  35%|███▍      | 165/472 [00:00<00:00, 2610.39it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.bias]
Loading weights:  35%|███▌      | 166/472 [00:00<00:00, 2624.31it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.weight]
Loading weights:  35%|███▌      | 166/472 [00:00<00:00, 2623.31it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.weight]
Loading weights:  35%|███▌      | 167/472 [00:00<00:00, 2637.49it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.bias]
Loading weights:  35%|███▌      | 167/472 [00:00<00:00, 2636.49it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.bias]
Loading weights:  36%|███▌      | 168/472 [00:00<00:00, 2650.64it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.weight]
Loading weights:  36%|███▌      | 168/472 [00:00<00:00, 2649.64it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.weight]
Loading weights:  36%|███▌      | 169/472 [00:00<00:00, 2663.43it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.bias]  
Loading weights:  36%|███▌      | 169/472 [00:00<00:00, 2662.42it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.bias]
Loading weights:  36%|███▌      | 170/472 [00:00<00:00, 2676.45it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.weight]
Loading weights:  36%|███▌      | 170/472 [00:00<00:00, 2675.40it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.weight]
Loading weights:  36%|███▌      | 171/472 [00:00<00:00, 2689.44it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  36%|███▌      | 171/472 [00:00<00:00, 2688.34it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  36%|███▋      | 172/472 [00:00<00:00, 2702.24it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  36%|███▋      | 172/472 [00:00<00:00, 2701.18it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  37%|███▋      | 173/472 [00:00<00:00, 2715.18it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.bias]      
Loading weights:  37%|███▋      | 173/472 [00:00<00:00, 2714.11it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.bias]
Loading weights:  37%|███▋      | 174/472 [00:00<00:00, 2728.14it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  37%|███▋      | 174/472 [00:00<00:00, 2727.11it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  37%|███▋      | 175/472 [00:00<00:00, 2741.11it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.bias]      
Loading weights:  37%|███▋      | 175/472 [00:00<00:00, 2740.08it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.bias]
Loading weights:  37%|███▋      | 176/472 [00:00<00:00, 2753.83it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  37%|███▋      | 176/472 [00:00<00:00, 2752.79it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  38%|███▊      | 177/472 [00:00<00:00, 2766.62it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  38%|███▊      | 177/472 [00:00<00:00, 2765.59it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  38%|███▊      | 178/472 [00:00<00:00, 2779.35it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  38%|███▊      | 178/472 [00:00<00:00, 2778.30it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  38%|███▊      | 179/472 [00:00<00:00, 2792.21it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.bias]  
Loading weights:  38%|███▊      | 179/472 [00:00<00:00, 2791.17it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.bias]
Loading weights:  38%|███▊      | 180/472 [00:00<00:00, 2805.04it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  38%|███▊      | 180/472 [00:00<00:00, 2804.05it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  38%|███▊      | 181/472 [00:00<00:00, 2817.78it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.bias]         
Loading weights:  38%|███▊      | 181/472 [00:00<00:00, 2816.60it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.bias]
Loading weights:  39%|███▊      | 182/472 [00:00<00:00, 2830.40it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.weight]
Loading weights:  39%|███▊      | 182/472 [00:00<00:00, 2829.35it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.weight]
Loading weights:  39%|███▉      | 183/472 [00:00<00:00, 2843.25it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.bias]    
Loading weights:  39%|███▉      | 183/472 [00:00<00:00, 2842.24it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.bias]
Loading weights:  39%|███▉      | 184/472 [00:00<00:00, 2856.08it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  39%|███▉      | 184/472 [00:00<00:00, 2855.06it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  39%|███▉      | 185/472 [00:00<00:00, 2868.84it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.bias]      
Loading weights:  39%|███▉      | 185/472 [00:00<00:00, 2867.82it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.bias]
Loading weights:  39%|███▉      | 186/472 [00:00<00:00, 2880.69it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.weight]
Loading weights:  39%|███▉      | 186/472 [00:00<00:00, 2879.62it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.weight]
Loading weights:  40%|███▉      | 187/472 [00:00<00:00, 2893.40it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  40%|███▉      | 187/472 [00:00<00:00, 2892.32it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  40%|███▉      | 188/472 [00:00<00:00, 2905.86it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  40%|███▉      | 188/472 [00:00<00:00, 2904.78it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  40%|████      | 189/472 [00:00<00:00, 2918.48it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.bias]      
Loading weights:  40%|████      | 189/472 [00:00<00:00, 2917.34it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.bias]
Loading weights:  40%|████      | 190/472 [00:00<00:00, 2930.98it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.weight]
Loading weights:  40%|████      | 190/472 [00:00<00:00, 2929.89it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.weight]
Loading weights:  40%|████      | 191/472 [00:00<00:00, 2943.49it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.bias]      
Loading weights:  40%|████      | 191/472 [00:00<00:00, 2942.36it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.bias]
Loading weights:  41%|████      | 192/472 [00:00<00:00, 2955.75it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.weight]
Loading weights:  41%|████      | 192/472 [00:00<00:00, 2954.68it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.weight]
Loading weights:  41%|████      | 193/472 [00:00<00:00, 2968.27it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.bias]
Loading weights:  41%|████      | 193/472 [00:00<00:00, 2967.10it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.bias]
Loading weights:  41%|████      | 194/472 [00:00<00:00, 2980.62it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.weight]
Loading weights:  41%|████      | 194/472 [00:00<00:00, 2979.03it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.weight]
Loading weights:  41%|████▏     | 195/472 [00:00<00:00, 2992.54it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.bias]  
Loading weights:  41%|████▏     | 195/472 [00:00<00:00, 2991.41it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.bias]
Loading weights:  42%|████▏     | 196/472 [00:00<00:00, 3004.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.weight]
Loading weights:  42%|████▏     | 196/472 [00:00<00:00, 3003.46it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.weight]
Loading weights:  42%|████▏     | 197/472 [00:00<00:00, 3016.80it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  42%|████▏     | 197/472 [00:00<00:00, 3015.60it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  42%|████▏     | 198/472 [00:00<00:00, 3028.82it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  42%|████▏     | 198/472 [00:00<00:00, 3027.66it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  42%|████▏     | 199/472 [00:00<00:00, 3040.96it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.bias]      
Loading weights:  42%|████▏     | 199/472 [00:00<00:00, 3039.75it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.bias]
Loading weights:  42%|████▏     | 200/472 [00:00<00:00, 3053.17it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  42%|████▏     | 200/472 [00:00<00:00, 3051.99it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  43%|████▎     | 201/472 [00:00<00:00, 3065.34it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.bias]      
Loading weights:  43%|████▎     | 201/472 [00:00<00:00, 3064.07it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.bias]
Loading weights:  43%|████▎     | 202/472 [00:00<00:00, 3077.29it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  43%|████▎     | 202/472 [00:00<00:00, 3076.13it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  43%|████▎     | 203/472 [00:00<00:00, 3089.52it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  43%|████▎     | 203/472 [00:00<00:00, 3088.31it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  43%|████▎     | 204/472 [00:00<00:00, 3101.49it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  43%|████▎     | 204/472 [00:00<00:00, 3100.35it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  43%|████▎     | 205/472 [00:00<00:00, 3113.76it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.bias]  
Loading weights:  43%|████▎     | 205/472 [00:00<00:00, 3112.62it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.bias]
Loading weights:  44%|████▎     | 206/472 [00:00<00:00, 3125.81it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  44%|████▎     | 206/472 [00:00<00:00, 3124.40it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  44%|████▍     | 207/472 [00:00<00:00, 3137.27it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.bias]         
Loading weights:  44%|████▍     | 207/472 [00:00<00:00, 3136.13it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.bias]
Loading weights:  44%|████▍     | 208/472 [00:00<00:00, 3149.27it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.weight]
Loading weights:  44%|████▍     | 208/472 [00:00<00:00, 3148.18it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.weight]
Loading weights:  44%|████▍     | 209/472 [00:00<00:00, 3161.16it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.bias]    
Loading weights:  44%|████▍     | 209/472 [00:00<00:00, 3160.01it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.bias]
Loading weights:  44%|████▍     | 210/472 [00:00<00:00, 3173.16it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  44%|████▍     | 210/472 [00:00<00:00, 3172.02it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  45%|████▍     | 211/472 [00:00<00:00, 3185.30it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.bias]      
Loading weights:  45%|████▍     | 211/472 [00:00<00:00, 3184.09it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.bias]
Loading weights:  45%|████▍     | 212/472 [00:00<00:00, 3197.35it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.weight]
Loading weights:  45%|████▍     | 212/472 [00:00<00:00, 3196.29it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.weight]
Loading weights:  45%|████▌     | 213/472 [00:00<00:00, 3209.52it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  45%|████▌     | 213/472 [00:00<00:00, 3208.42it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  45%|████▌     | 214/472 [00:00<00:00, 3221.34it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  45%|████▌     | 214/472 [00:00<00:00, 3220.08it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  46%|████▌     | 215/472 [00:00<00:00, 3232.70it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.bias]      
Loading weights:  46%|████▌     | 215/472 [00:00<00:00, 3231.53it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.bias]
Loading weights:  46%|████▌     | 216/472 [00:00<00:00, 3244.52it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.weight]
Loading weights:  46%|████▌     | 216/472 [00:00<00:00, 3243.34it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.weight]
Loading weights:  46%|████▌     | 217/472 [00:00<00:00, 3256.41it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.bias]      
Loading weights:  46%|████▌     | 217/472 [00:00<00:00, 3255.24it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.bias]
Loading weights:  46%|████▌     | 218/472 [00:00<00:00, 3268.38it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.weight]
Loading weights:  46%|████▌     | 218/472 [00:00<00:00, 3267.25it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.weight]
Loading weights:  46%|████▋     | 219/472 [00:00<00:00, 3280.36it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.bias]
Loading weights:  46%|████▋     | 219/472 [00:00<00:00, 3278.94it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.bias]
Loading weights:  47%|████▋     | 220/472 [00:00<00:00, 3292.00it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.weight]
Loading weights:  47%|████▋     | 220/472 [00:00<00:00, 3290.81it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.weight]
Loading weights:  47%|████▋     | 221/472 [00:00<00:00, 3303.90it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.bias]  
Loading weights:  47%|████▋     | 221/472 [00:00<00:00, 3302.45it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.bias]
Loading weights:  47%|████▋     | 222/472 [00:00<00:00, 3315.36it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.weight]
Loading weights:  47%|████▋     | 222/472 [00:00<00:00, 3314.23it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.weight]
Loading weights:  47%|████▋     | 223/472 [00:00<00:00, 3326.92it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  47%|████▋     | 223/472 [00:00<00:00, 3325.69it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  47%|████▋     | 224/472 [00:00<00:00, 3338.36it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  47%|████▋     | 224/472 [00:00<00:00, 3337.11it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  48%|████▊     | 225/472 [00:00<00:00, 3349.96it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.bias]      
Loading weights:  48%|████▊     | 225/472 [00:00<00:00, 3348.76it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.bias]
Loading weights:  48%|████▊     | 226/472 [00:00<00:00, 3361.70it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  48%|████▊     | 226/472 [00:00<00:00, 3360.44it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  48%|████▊     | 227/472 [00:00<00:00, 3373.20it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.bias]      
Loading weights:  48%|████▊     | 227/472 [00:00<00:00, 3372.07it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.bias]
Loading weights:  48%|████▊     | 228/472 [00:00<00:00, 3384.95it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  48%|████▊     | 228/472 [00:00<00:00, 3383.74it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  49%|████▊     | 229/472 [00:00<00:00, 3396.41it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  49%|████▊     | 229/472 [00:00<00:00, 3395.21it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  49%|████▊     | 230/472 [00:00<00:00, 3407.97it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  49%|████▊     | 230/472 [00:00<00:00, 3406.81it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  49%|████▉     | 231/472 [00:00<00:00, 3419.55it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.bias]  
Loading weights:  49%|████▉     | 231/472 [00:00<00:00, 3418.22it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.bias]
Loading weights:  49%|████▉     | 232/472 [00:00<00:00, 3430.89it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  49%|████▉     | 232/472 [00:00<00:00, 3429.73it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  49%|████▉     | 233/472 [00:00<00:00, 3442.37it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.bias]         
Loading weights:  49%|████▉     | 233/472 [00:00<00:00, 3441.15it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.bias]
Loading weights:  50%|████▉     | 234/472 [00:00<00:00, 3453.83it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.weight]
Loading weights:  50%|████▉     | 234/472 [00:00<00:00, 3452.60it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.weight]
Loading weights:  50%|████▉     | 235/472 [00:00<00:00, 3465.10it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.bias]    
Loading weights:  50%|████▉     | 235/472 [00:00<00:00, 3463.94it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.bias]
Loading weights:  50%|█████     | 236/472 [00:00<00:00, 3476.42it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  50%|█████     | 236/472 [00:00<00:00, 3475.19it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  50%|█████     | 237/472 [00:00<00:00, 3487.76it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.bias]      
Loading weights:  50%|█████     | 237/472 [00:00<00:00, 3486.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.bias]
Loading weights:  50%|█████     | 238/472 [00:00<00:00, 3499.39it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.weight]
Loading weights:  50%|█████     | 238/472 [00:00<00:00, 3498.25it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.weight]
Loading weights:  51%|█████     | 239/472 [00:00<00:00, 3510.94it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  51%|█████     | 239/472 [00:00<00:00, 3509.75it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  51%|█████     | 240/472 [00:00<00:00, 3522.21it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  51%|█████     | 240/472 [00:00<00:00, 3520.98it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  51%|█████     | 241/472 [00:00<00:00, 3533.47it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.bias]      
Loading weights:  51%|█████     | 241/472 [00:00<00:00, 3532.18it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.bias]
Loading weights:  51%|█████▏    | 242/472 [00:00<00:00, 3544.70it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.weight]
Loading weights:  51%|█████▏    | 242/472 [00:00<00:00, 3543.45it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.weight]
Loading weights:  51%|█████▏    | 243/472 [00:00<00:00, 3556.02it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.bias]      
Loading weights:  51%|█████▏    | 243/472 [00:00<00:00, 3554.82it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.bias]
Loading weights:  52%|█████▏    | 244/472 [00:00<00:00, 3567.42it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.weight]
Loading weights:  52%|█████▏    | 244/472 [00:00<00:00, 3566.16it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.weight]
Loading weights:  52%|█████▏    | 245/472 [00:00<00:00, 3578.63it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.bias]
Loading weights:  52%|█████▏    | 245/472 [00:00<00:00, 3577.28it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.bias]
Loading weights:  52%|█████▏    | 246/472 [00:00<00:00, 3589.73it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.weight]
Loading weights:  52%|█████▏    | 246/472 [00:00<00:00, 3588.52it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.weight]
Loading weights:  52%|█████▏    | 247/472 [00:00<00:00, 3600.95it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.bias]  
Loading weights:  52%|█████▏    | 247/472 [00:00<00:00, 3599.69it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.bias]
Loading weights:  53%|█████▎    | 248/472 [00:00<00:00, 3612.21it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.weight]
Loading weights:  53%|█████▎    | 248/472 [00:00<00:00, 3610.95it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.weight]
Loading weights:  53%|█████▎    | 249/472 [00:00<00:00, 3623.13it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  53%|█████▎    | 249/472 [00:00<00:00, 3621.87it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  53%|█████▎    | 250/472 [00:00<00:00, 3633.99it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  53%|█████▎    | 250/472 [00:00<00:00, 3632.66it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  53%|█████▎    | 251/472 [00:00<00:00, 3644.97it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.bias]      
Loading weights:  53%|█████▎    | 251/472 [00:00<00:00, 3643.71it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.bias]
Loading weights:  53%|█████▎    | 252/472 [00:00<00:00, 3656.09it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  53%|█████▎    | 252/472 [00:00<00:00, 3654.78it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  54%|█████▎    | 253/472 [00:00<00:00, 3667.00it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.bias]      
Loading weights:  54%|█████▎    | 253/472 [00:00<00:00, 3665.67it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.bias]
Loading weights:  54%|█████▍    | 254/472 [00:00<00:00, 3677.92it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  54%|█████▍    | 254/472 [00:00<00:00, 3676.64it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  54%|█████▍    | 255/472 [00:00<00:00, 3688.97it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  54%|█████▍    | 255/472 [00:00<00:00, 3687.69it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  54%|█████▍    | 256/472 [00:00<00:00, 3699.96it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  54%|█████▍    | 256/472 [00:00<00:00, 3698.62it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  54%|█████▍    | 257/472 [00:00<00:00, 3710.87it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.bias]  
Loading weights:  54%|█████▍    | 257/472 [00:00<00:00, 3709.49it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.bias]
Loading weights:  55%|█████▍    | 258/472 [00:00<00:00, 3721.67it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  55%|█████▍    | 258/472 [00:00<00:00, 3720.43it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  55%|█████▍    | 259/472 [00:00<00:00, 3732.79it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.bias]         
Loading weights:  55%|█████▍    | 259/472 [00:00<00:00, 3731.51it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.bias]
Loading weights:  55%|█████▌    | 260/472 [00:00<00:00, 3743.86it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.weight]
Loading weights:  55%|█████▌    | 260/472 [00:00<00:00, 3742.51it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.weight]
Loading weights:  55%|█████▌    | 261/472 [00:00<00:00, 3754.69it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.bias]    
Loading weights:  55%|█████▌    | 261/472 [00:00<00:00, 3753.45it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.bias]
Loading weights:  56%|█████▌    | 262/472 [00:00<00:00, 3765.78it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  56%|█████▌    | 262/472 [00:00<00:00, 3764.58it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  56%|█████▌    | 263/472 [00:00<00:00, 3776.96it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.bias]      
Loading weights:  56%|█████▌    | 263/472 [00:00<00:00, 3775.70it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.bias]
Loading weights:  56%|█████▌    | 264/472 [00:00<00:00, 3788.05it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.weight]
Loading weights:  56%|█████▌    | 264/472 [00:00<00:00, 3786.79it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.weight]
Loading weights:  56%|█████▌    | 265/472 [00:00<00:00, 3799.12it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  56%|█████▌    | 265/472 [00:00<00:00, 3797.81it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  56%|█████▋    | 266/472 [00:00<00:00, 3809.68it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  56%|█████▋    | 266/472 [00:00<00:00, 3808.38it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  57%|█████▋    | 267/472 [00:00<00:00, 3820.56it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.bias]      
Loading weights:  57%|█████▋    | 267/472 [00:00<00:00, 3819.21it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.bias]
Loading weights:  57%|█████▋    | 268/472 [00:00<00:00, 3831.30it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.weight]
Loading weights:  57%|█████▋    | 268/472 [00:00<00:00, 3829.94it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.weight]
Loading weights:  57%|█████▋    | 269/472 [00:00<00:00, 3841.93it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.bias]      
Loading weights:  57%|█████▋    | 269/472 [00:00<00:00, 3840.61it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.bias]
Loading weights:  57%|█████▋    | 270/472 [00:00<00:00, 3852.58it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.weight]
Loading weights:  57%|█████▋    | 270/472 [00:00<00:00, 3851.32it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.weight]
Loading weights:  57%|█████▋    | 271/472 [00:00<00:00, 3863.48it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.bias]
Loading weights:  57%|█████▋    | 271/472 [00:00<00:00, 3862.17it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.bias]
Loading weights:  58%|█████▊    | 272/472 [00:00<00:00, 3874.20it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.weight]
Loading weights:  58%|█████▊    | 272/472 [00:00<00:00, 3872.88it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.weight]
Loading weights:  58%|█████▊    | 273/472 [00:00<00:00, 3884.85it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.bias]  
Loading weights:  58%|█████▊    | 273/472 [00:00<00:00, 3883.54it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.bias]
Loading weights:  58%|█████▊    | 274/472 [00:00<00:00, 3895.65it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.weight]
Loading weights:  58%|█████▊    | 274/472 [00:00<00:00, 3894.38it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.weight]
Loading weights:  58%|█████▊    | 275/472 [00:00<00:00, 3906.47it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  58%|█████▊    | 275/472 [00:00<00:00, 3905.19it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  58%|█████▊    | 276/472 [00:00<00:00, 3917.06it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  58%|█████▊    | 276/472 [00:00<00:00, 3915.63it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  59%|█████▊    | 277/472 [00:00<00:00, 3927.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.bias]      
Loading weights:  59%|█████▊    | 277/472 [00:00<00:00, 3926.31it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.bias]
Loading weights:  59%|█████▉    | 278/472 [00:00<00:00, 3938.23it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  59%|█████▉    | 278/472 [00:00<00:00, 3936.94it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  59%|█████▉    | 279/472 [00:00<00:00, 3948.71it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.bias]      
Loading weights:  59%|█████▉    | 279/472 [00:00<00:00, 3947.44it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.bias]
Loading weights:  59%|█████▉    | 280/472 [00:00<00:00, 3959.33it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  59%|█████▉    | 280/472 [00:00<00:00, 3958.10it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  60%|█████▉    | 281/472 [00:00<00:00, 3969.94it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  60%|█████▉    | 281/472 [00:00<00:00, 3968.60it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  60%|█████▉    | 282/472 [00:00<00:00, 3980.25it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  60%|█████▉    | 282/472 [00:00<00:00, 3978.89it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  60%|█████▉    | 283/472 [00:00<00:00, 3990.76it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.bias]  
Loading weights:  60%|█████▉    | 283/472 [00:00<00:00, 3989.41it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.bias]
Loading weights:  60%|██████    | 284/472 [00:00<00:00, 4001.08it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  60%|██████    | 284/472 [00:00<00:00, 3999.66it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  60%|██████    | 285/472 [00:00<00:00, 4011.37it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.bias]         
Loading weights:  60%|██████    | 285/472 [00:00<00:00, 4010.01it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.bias]
Loading weights:  61%|██████    | 286/472 [00:00<00:00, 4021.71it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.weight]
Loading weights:  61%|██████    | 286/472 [00:00<00:00, 4020.43it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.weight]
Loading weights:  61%|██████    | 287/472 [00:00<00:00, 4032.15it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.bias]    
Loading weights:  61%|██████    | 287/472 [00:00<00:00, 4029.70it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.bias]
Loading weights:  61%|██████    | 288/472 [00:00<00:00, 4040.17it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  61%|██████    | 288/472 [00:00<00:00, 4038.65it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  61%|██████    | 289/472 [00:00<00:00, 4049.95it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.bias]      
Loading weights:  61%|██████    | 289/472 [00:00<00:00, 4048.25it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.bias]
Loading weights:  61%|██████▏   | 290/472 [00:00<00:00, 4059.87it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.weight]
Loading weights:  61%|██████▏   | 290/472 [00:00<00:00, 4058.44it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.weight]
Loading weights:  62%|██████▏   | 291/472 [00:00<00:00, 4069.89it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  62%|██████▏   | 291/472 [00:00<00:00, 4068.46it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  62%|██████▏   | 292/472 [00:00<00:00, 4079.87it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  62%|██████▏   | 292/472 [00:00<00:00, 4078.16it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  62%|██████▏   | 293/472 [00:00<00:00, 4089.61it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.bias]      
Loading weights:  62%|██████▏   | 293/472 [00:00<00:00, 4088.23it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.bias]
Loading weights:  62%|██████▏   | 294/472 [00:00<00:00, 4099.21it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.weight]
Loading weights:  62%|██████▏   | 294/472 [00:00<00:00, 4097.91it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.weight]
Loading weights:  62%|██████▎   | 295/472 [00:00<00:00, 4109.15it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.bias]      
Loading weights:  62%|██████▎   | 295/472 [00:00<00:00, 4107.72it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.bias]
Loading weights:  63%|██████▎   | 296/472 [00:00<00:00, 4119.12it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.weight]
Loading weights:  63%|██████▎   | 296/472 [00:00<00:00, 4117.75it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.weight]
Loading weights:  63%|██████▎   | 297/472 [00:00<00:00, 4129.25it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.bias]
Loading weights:  63%|██████▎   | 297/472 [00:00<00:00, 4127.87it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.bias]
Loading weights:  63%|██████▎   | 298/472 [00:00<00:00, 4139.42it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.weight]
Loading weights:  63%|██████▎   | 298/472 [00:00<00:00, 4138.04it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.weight]
Loading weights:  63%|██████▎   | 299/472 [00:00<00:00, 4149.38it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.bias]  
Loading weights:  63%|██████▎   | 299/472 [00:00<00:00, 4147.94it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.bias]
Loading weights:  64%|██████▎   | 300/472 [00:00<00:00, 4159.50it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.weight]
Loading weights:  64%|██████▎   | 300/472 [00:00<00:00, 4158.11it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.weight]
Loading weights:  64%|██████▍   | 301/472 [00:00<00:00, 4169.38it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  64%|██████▍   | 301/472 [00:00<00:00, 4168.06it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  64%|██████▍   | 302/472 [00:00<00:00, 4179.54it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  64%|██████▍   | 302/472 [00:00<00:00, 4178.09it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  64%|██████▍   | 303/472 [00:00<00:00, 4189.37it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.bias]      
Loading weights:  64%|██████▍   | 303/472 [00:00<00:00, 4187.97it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.bias]
Loading weights:  64%|██████▍   | 304/472 [00:00<00:00, 4199.30it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  64%|██████▍   | 304/472 [00:00<00:00, 4197.69it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  65%|██████▍   | 305/472 [00:00<00:00, 4208.93it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.bias]      
Loading weights:  65%|██████▍   | 305/472 [00:00<00:00, 4207.31it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.bias]
Loading weights:  65%|██████▍   | 306/472 [00:00<00:00, 4218.83it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  65%|██████▍   | 306/472 [00:00<00:00, 4217.32it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  65%|██████▌   | 307/472 [00:00<00:00, 4228.71it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  65%|██████▌   | 307/472 [00:00<00:00, 4227.38it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  65%|██████▌   | 308/472 [00:00<00:00, 4238.58it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  65%|██████▌   | 308/472 [00:00<00:00, 4237.17it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  65%|██████▌   | 309/472 [00:00<00:00, 4248.36it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.bias]  
Loading weights:  65%|██████▌   | 309/472 [00:00<00:00, 4246.97it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.bias]
Loading weights:  66%|██████▌   | 310/472 [00:00<00:00, 4257.55it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  66%|██████▌   | 310/472 [00:00<00:00, 4256.03it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  66%|██████▌   | 311/472 [00:00<00:00, 4267.54it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.bias]         
Loading weights:  66%|██████▌   | 311/472 [00:00<00:00, 4266.06it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.bias]
Loading weights:  66%|██████▌   | 312/472 [00:00<00:00, 4277.50it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.weight]
Loading weights:  66%|██████▌   | 312/472 [00:00<00:00, 4276.15it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.weight]
Loading weights:  66%|██████▋   | 313/472 [00:00<00:00, 4287.39it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.bias]    
Loading weights:  66%|██████▋   | 313/472 [00:00<00:00, 4286.04it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.bias]
Loading weights:  67%|██████▋   | 314/472 [00:00<00:00, 4297.32it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  67%|██████▋   | 314/472 [00:00<00:00, 4295.96it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  67%|██████▋   | 315/472 [00:00<00:00, 4307.33it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.bias]      
Loading weights:  67%|██████▋   | 315/472 [00:00<00:00, 4306.00it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.bias]
Loading weights:  67%|██████▋   | 316/472 [00:00<00:00, 4317.19it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.weight]
Loading weights:  67%|██████▋   | 316/472 [00:00<00:00, 4315.70it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.weight]
Loading weights:  67%|██████▋   | 317/472 [00:00<00:00, 4327.11it/s, Materializing param=text_decoder.cls.predictions.bias]                     
Loading weights:  67%|██████▋   | 317/472 [00:00<00:00, 4325.64it/s, Materializing param=text_decoder.cls.predictions.bias]
Loading weights:  67%|██████▋   | 318/472 [00:00<00:00, 4337.16it/s, Materializing param=text_decoder.cls.predictions.decoder.weight]
Loading weights:  67%|██████▋   | 318/472 [00:00<00:00, 4335.27it/s, Materializing param=text_decoder.cls.predictions.decoder.weight]
Loading weights:  68%|██████▊   | 319/472 [00:00<00:00, 4346.53it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.bias]
Loading weights:  68%|██████▊   | 319/472 [00:00<00:00, 4345.16it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.bias]
Loading weights:  68%|██████▊   | 320/472 [00:00<00:00, 4356.60it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.weight]
Loading weights:  68%|██████▊   | 320/472 [00:00<00:00, 4354.99it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.weight]
Loading weights:  68%|██████▊   | 321/472 [00:00<00:00, 4366.40it/s, Materializing param=text_decoder.cls.predictions.transform.dense.bias]      
Loading weights:  68%|██████▊   | 321/472 [00:00<00:00, 4364.80it/s, Materializing param=text_decoder.cls.predictions.transform.dense.bias]
Loading weights:  68%|██████▊   | 322/472 [00:00<00:00, 4375.78it/s, Materializing param=text_decoder.cls.predictions.transform.dense.weight]
Loading weights:  68%|██████▊   | 322/472 [00:00<00:00, 4374.30it/s, Materializing param=text_decoder.cls.predictions.transform.dense.weight]
Loading weights:  68%|██████▊   | 323/472 [00:00<00:00, 4385.80it/s, Materializing param=vision_model.embeddings.class_embedding]            
Loading weights:  68%|██████▊   | 323/472 [00:00<00:00, 4384.42it/s, Materializing param=vision_model.embeddings.class_embedding]
Loading weights:  69%|██████▊   | 324/472 [00:00<00:00, 4393.93it/s, Materializing param=vision_model.embeddings.patch_embedding.bias]
Loading weights:  69%|██████▊   | 324/472 [00:00<00:00, 4392.62it/s, Materializing param=vision_model.embeddings.patch_embedding.bias]
Loading weights:  69%|██████▉   | 325/472 [00:00<00:00, 4403.21it/s, Materializing param=vision_model.embeddings.patch_embedding.weight]
Loading weights:  69%|██████▉   | 325/472 [00:00<00:00, 4401.88it/s, Materializing param=vision_model.embeddings.patch_embedding.weight]
Loading weights:  69%|██████▉   | 326/472 [00:00<00:00, 4413.23it/s, Materializing param=vision_model.embeddings.position_embedding]    
Loading weights:  69%|██████▉   | 326/472 [00:00<00:00, 4411.79it/s, Materializing param=vision_model.embeddings.position_embedding]
Loading weights:  69%|██████▉   | 327/472 [00:00<00:00, 4423.16it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.bias]
Loading weights:  69%|██████▉   | 327/472 [00:00<00:00, 4421.85it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.bias]
Loading weights:  69%|██████▉   | 328/472 [00:00<00:00, 4432.31it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.weight]
Loading weights:  69%|██████▉   | 328/472 [00:00<00:00, 4430.87it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.weight]
Loading weights:  70%|██████▉   | 329/472 [00:00<00:00, 4441.76it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.bias]  
Loading weights:  70%|██████▉   | 329/472 [00:00<00:00, 4440.43it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.bias]
Loading weights:  70%|██████▉   | 330/472 [00:00<00:00, 4451.46it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.weight]
Loading weights:  70%|██████▉   | 330/472 [00:00<00:00, 4450.09it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.weight]
Loading weights:  70%|███████   | 331/472 [00:00<00:00, 4461.28it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.bias]      
Loading weights:  70%|███████   | 331/472 [00:00<00:00, 4460.02it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.bias]
Loading weights:  70%|███████   | 332/472 [00:00<00:00, 4470.97it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.weight]
Loading weights:  70%|███████   | 332/472 [00:00<00:00, 4469.64it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.weight]
Loading weights:  71%|███████   | 333/472 [00:00<00:00, 4480.50it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.bias]  
Loading weights:  71%|███████   | 333/472 [00:00<00:00, 4479.05it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.bias]
Loading weights:  71%|███████   | 334/472 [00:00<00:00, 4489.74it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.weight]
Loading weights:  71%|███████   | 334/472 [00:00<00:00, 4488.34it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.weight]
Loading weights:  71%|███████   | 335/472 [00:00<00:00, 4499.61it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.bias]
Loading weights:  71%|███████   | 335/472 [00:00<00:00, 4498.21it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.bias]
Loading weights:  71%|███████   | 336/472 [00:00<00:00, 4508.92it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.weight]
Loading weights:  71%|███████   | 336/472 [00:00<00:00, 4507.41it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.weight]
Loading weights:  71%|███████▏  | 337/472 [00:00<00:00, 4518.03it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.bias]         
Loading weights:  71%|███████▏  | 337/472 [00:00<00:00, 4516.59it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.bias]
Loading weights:  72%|███████▏  | 338/472 [00:00<00:00, 4527.87it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.weight]
Loading weights:  72%|███████▏  | 338/472 [00:00<00:00, 4526.59it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.weight]
Loading weights:  72%|███████▏  | 339/472 [00:00<00:00, 4537.49it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.bias]    
Loading weights:  72%|███████▏  | 339/472 [00:00<00:00, 4536.03it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.bias]
Loading weights:  72%|███████▏  | 340/472 [00:00<00:00, 4546.56it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.weight]
Loading weights:  72%|███████▏  | 340/472 [00:00<00:00, 4545.11it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.weight]
Loading weights:  72%|███████▏  | 341/472 [00:00<00:00, 4556.15it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.bias]  
Loading weights:  72%|███████▏  | 341/472 [00:00<00:00, 4554.63it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.bias]
Loading weights:  72%|███████▏  | 342/472 [00:00<00:00, 4565.35it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.weight]
Loading weights:  72%|███████▏  | 342/472 [00:00<00:00, 4563.90it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.weight]
Loading weights:  73%|███████▎  | 343/472 [00:00<00:00, 4574.92it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.bias]      
Loading weights:  73%|███████▎  | 343/472 [00:00<00:00, 4573.52it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.bias]
Loading weights:  73%|███████▎  | 344/472 [00:00<00:00, 4584.22it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.weight]
Loading weights:  73%|███████▎  | 344/472 [00:00<00:00, 4582.83it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.weight]
Loading weights:  73%|███████▎  | 345/472 [00:00<00:00, 4593.69it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.bias]  
Loading weights:  73%|███████▎  | 345/472 [00:00<00:00, 4592.30it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.bias]
Loading weights:  73%|███████▎  | 346/472 [00:00<00:00, 4602.91it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.weight]
Loading weights:  73%|███████▎  | 346/472 [00:00<00:00, 4601.44it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.weight]
Loading weights:  74%|███████▎  | 347/472 [00:00<00:00, 4611.97it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.bias]
Loading weights:  74%|███████▎  | 347/472 [00:00<00:00, 4610.56it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.bias]
Loading weights:  74%|███████▎  | 348/472 [00:00<00:00, 4621.15it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.weight]
Loading weights:  74%|███████▎  | 348/472 [00:00<00:00, 4619.75it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.weight]
Loading weights:  74%|███████▍  | 349/472 [00:00<00:00, 4630.56it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.bias]         
Loading weights:  74%|███████▍  | 349/472 [00:00<00:00, 4629.20it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.bias]
Loading weights:  74%|███████▍  | 350/472 [00:00<00:00, 4639.82it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.weight]
Loading weights:  74%|███████▍  | 350/472 [00:00<00:00, 4638.41it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.weight]
Loading weights:  74%|███████▍  | 351/472 [00:00<00:00, 4649.26it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.bias]    
Loading weights:  74%|███████▍  | 351/472 [00:00<00:00, 4647.83it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.bias]
Loading weights:  75%|███████▍  | 352/472 [00:00<00:00, 4658.31it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.weight]
Loading weights:  75%|███████▍  | 352/472 [00:00<00:00, 4656.90it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.weight]
Loading weights:  75%|███████▍  | 353/472 [00:00<00:00, 4667.89it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.bias]  
Loading weights:  75%|███████▍  | 353/472 [00:00<00:00, 4666.48it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.bias]
Loading weights:  75%|███████▌  | 354/472 [00:00<00:00, 4677.03it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.weight]
Loading weights:  75%|███████▌  | 354/472 [00:00<00:00, 4675.61it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.weight]
Loading weights:  75%|███████▌  | 355/472 [00:00<00:00, 4686.28it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.bias]      
Loading weights:  75%|███████▌  | 355/472 [00:00<00:00, 4684.85it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.bias]
Loading weights:  75%|███████▌  | 356/472 [00:00<00:00, 4695.27it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.weight]
Loading weights:  75%|███████▌  | 356/472 [00:00<00:00, 4693.86it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.weight]
Loading weights:  76%|███████▌  | 357/472 [00:00<00:00, 4704.50it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.bias]  
Loading weights:  76%|███████▌  | 357/472 [00:00<00:00, 4703.20it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.bias]
Loading weights:  76%|███████▌  | 358/472 [00:00<00:00, 4713.77it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.weight]
Loading weights:  76%|███████▌  | 358/472 [00:00<00:00, 4712.46it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.weight]
Loading weights:  76%|███████▌  | 359/472 [00:00<00:00, 4723.32it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.bias]
Loading weights:  76%|███████▌  | 359/472 [00:00<00:00, 4721.95it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.bias]
Loading weights:  76%|███████▋  | 360/472 [00:00<00:00, 4732.61it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.weight]
Loading weights:  76%|███████▋  | 360/472 [00:00<00:00, 4731.25it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.weight]
Loading weights:  76%|███████▋  | 361/472 [00:00<00:00, 4741.58it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.bias]         
Loading weights:  76%|███████▋  | 361/472 [00:00<00:00, 4740.14it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.bias]
Loading weights:  77%|███████▋  | 362/472 [00:00<00:00, 4750.54it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.weight]
Loading weights:  77%|███████▋  | 362/472 [00:00<00:00, 4749.11it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.weight]
Loading weights:  77%|███████▋  | 363/472 [00:00<00:00, 4759.59it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.bias]    
Loading weights:  77%|███████▋  | 363/472 [00:00<00:00, 4758.11it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.bias]
Loading weights:  77%|███████▋  | 364/472 [00:00<00:00, 4768.40it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.weight]
Loading weights:  77%|███████▋  | 364/472 [00:00<00:00, 4766.95it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.weight]
Loading weights:  77%|███████▋  | 365/472 [00:00<00:00, 4777.62it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.bias]  
Loading weights:  77%|███████▋  | 365/472 [00:00<00:00, 4776.11it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.bias]
Loading weights:  78%|███████▊  | 366/472 [00:00<00:00, 4786.45it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.weight]
Loading weights:  78%|███████▊  | 366/472 [00:00<00:00, 4784.76it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.weight]
Loading weights:  78%|███████▊  | 367/472 [00:00<00:00, 4795.13it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.bias]      
Loading weights:  78%|███████▊  | 367/472 [00:00<00:00, 4793.82it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.bias]
Loading weights:  78%|███████▊  | 368/472 [00:00<00:00, 4802.48it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.weight]
Loading weights:  78%|███████▊  | 368/472 [00:00<00:00, 4800.11it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.weight]
Loading weights:  78%|███████▊  | 369/472 [00:00<00:00, 4809.38it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.bias]  
Loading weights:  78%|███████▊  | 369/472 [00:00<00:00, 4807.50it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.bias]
Loading weights:  78%|███████▊  | 370/472 [00:00<00:00, 4817.52it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.weight]
Loading weights:  78%|███████▊  | 370/472 [00:00<00:00, 4816.03it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.weight]
Loading weights:  79%|███████▊  | 371/472 [00:00<00:00, 4826.21it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.bias]
Loading weights:  79%|███████▊  | 371/472 [00:00<00:00, 4824.58it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.bias]
Loading weights:  79%|███████▉  | 372/472 [00:00<00:00, 4834.63it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.weight]
Loading weights:  79%|███████▉  | 372/472 [00:00<00:00, 4833.12it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.weight]
Loading weights:  79%|███████▉  | 373/472 [00:00<00:00, 4843.40it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.bias]         
Loading weights:  79%|███████▉  | 373/472 [00:00<00:00, 4841.96it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.bias]
Loading weights:  79%|███████▉  | 374/472 [00:00<00:00, 4852.55it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.weight]
Loading weights:  79%|███████▉  | 374/472 [00:00<00:00, 4851.10it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.weight]
Loading weights:  79%|███████▉  | 375/472 [00:00<00:00, 4861.74it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.bias]    
Loading weights:  79%|███████▉  | 375/472 [00:00<00:00, 4860.35it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.bias]
Loading weights:  80%|███████▉  | 376/472 [00:00<00:00, 4870.61it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.weight]
Loading weights:  80%|███████▉  | 376/472 [00:00<00:00, 4869.21it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.weight]
Loading weights:  80%|███████▉  | 377/472 [00:00<00:00, 4879.75it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.bias]  
Loading weights:  80%|███████▉  | 377/472 [00:00<00:00, 4878.36it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.bias]
Loading weights:  80%|████████  | 378/472 [00:00<00:00, 4889.10it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.weight]
Loading weights:  80%|████████  | 378/472 [00:00<00:00, 4887.64it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.weight]
Loading weights:  80%|████████  | 379/472 [00:00<00:00, 4898.11it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.bias]      
Loading weights:  80%|████████  | 379/472 [00:00<00:00, 4896.66it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.bias]
Loading weights:  81%|████████  | 380/472 [00:00<00:00, 4907.02it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.weight]
Loading weights:  81%|████████  | 380/472 [00:00<00:00, 4905.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.weight]
Loading weights:  81%|████████  | 381/472 [00:00<00:00, 4915.88it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.bias]  
Loading weights:  81%|████████  | 381/472 [00:00<00:00, 4914.37it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.bias]
Loading weights:  81%|████████  | 382/472 [00:00<00:00, 4924.71it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.weight]
Loading weights:  81%|████████  | 382/472 [00:00<00:00, 4923.33it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.weight]
Loading weights:  81%|████████  | 383/472 [00:00<00:00, 4933.79it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.bias]
Loading weights:  81%|████████  | 383/472 [00:00<00:00, 4932.32it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.bias]
Loading weights:  81%|████████▏ | 384/472 [00:00<00:00, 4942.59it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.weight]
Loading weights:  81%|████████▏ | 384/472 [00:00<00:00, 4941.06it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.weight]
Loading weights:  82%|████████▏ | 385/472 [00:00<00:00, 4951.39it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.bias]         
Loading weights:  82%|████████▏ | 385/472 [00:00<00:00, 4949.92it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.bias]
Loading weights:  82%|████████▏ | 386/472 [00:00<00:00, 4960.36it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.weight]
Loading weights:  82%|████████▏ | 386/472 [00:00<00:00, 4958.90it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.weight]
Loading weights:  82%|████████▏ | 387/472 [00:00<00:00, 4969.13it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.bias]    
Loading weights:  82%|████████▏ | 387/472 [00:00<00:00, 4967.73it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.bias]
Loading weights:  82%|████████▏ | 388/472 [00:00<00:00, 4977.95it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.weight]
Loading weights:  82%|████████▏ | 388/472 [00:00<00:00, 4976.53it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.weight]
Loading weights:  82%|████████▏ | 389/472 [00:00<00:00, 4986.99it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.bias]  
Loading weights:  82%|████████▏ | 389/472 [00:00<00:00, 4985.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.bias]
Loading weights:  83%|████████▎ | 390/472 [00:00<00:00, 4995.90it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.weight]
Loading weights:  83%|████████▎ | 390/472 [00:00<00:00, 4994.44it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.weight]
Loading weights:  83%|████████▎ | 391/472 [00:00<00:00, 5004.82it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.bias]      
Loading weights:  83%|████████▎ | 391/472 [00:00<00:00, 5003.33it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.bias]
Loading weights:  83%|████████▎ | 392/472 [00:00<00:00, 5013.82it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.weight]
Loading weights:  83%|████████▎ | 392/472 [00:00<00:00, 5012.40it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.weight]
Loading weights:  83%|████████▎ | 393/472 [00:00<00:00, 5022.75it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.bias]  
Loading weights:  83%|████████▎ | 393/472 [00:00<00:00, 5021.34it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.bias]
Loading weights:  83%|████████▎ | 394/472 [00:00<00:00, 5031.87it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.weight]
Loading weights:  83%|████████▎ | 394/472 [00:00<00:00, 5030.38it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.weight]
Loading weights:  84%|████████▎ | 395/472 [00:00<00:00, 5040.71it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.bias]
Loading weights:  84%|████████▎ | 395/472 [00:00<00:00, 5039.36it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.bias]
Loading weights:  84%|████████▍ | 396/472 [00:00<00:00, 5049.66it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.weight]
Loading weights:  84%|████████▍ | 396/472 [00:00<00:00, 5048.25it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.weight]
Loading weights:  84%|████████▍ | 397/472 [00:00<00:00, 5058.61it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.bias]         
Loading weights:  84%|████████▍ | 397/472 [00:00<00:00, 5057.06it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.bias]
Loading weights:  84%|████████▍ | 398/472 [00:00<00:00, 5067.49it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.weight]
Loading weights:  84%|████████▍ | 398/472 [00:00<00:00, 5066.01it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.weight]
Loading weights:  85%|████████▍ | 399/472 [00:00<00:00, 5076.34it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.bias]    
Loading weights:  85%|████████▍ | 399/472 [00:00<00:00, 5074.74it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.bias]
Loading weights:  85%|████████▍ | 400/472 [00:00<00:00, 5084.99it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.weight]
Loading weights:  85%|████████▍ | 400/472 [00:00<00:00, 5083.50it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.weight]
Loading weights:  85%|████████▍ | 401/472 [00:00<00:00, 5093.75it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.bias]  
Loading weights:  85%|████████▍ | 401/472 [00:00<00:00, 5092.26it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.bias]
Loading weights:  85%|████████▌ | 402/472 [00:00<00:00, 5102.44it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.weight]
Loading weights:  85%|████████▌ | 402/472 [00:00<00:00, 5100.95it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.weight]
Loading weights:  85%|████████▌ | 403/472 [00:00<00:00, 5111.09it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.bias]      
Loading weights:  85%|████████▌ | 403/472 [00:00<00:00, 5109.69it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.bias]
Loading weights:  86%|████████▌ | 404/472 [00:00<00:00, 5119.89it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.weight]
Loading weights:  86%|████████▌ | 404/472 [00:00<00:00, 5118.47it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.weight]
Loading weights:  86%|████████▌ | 405/472 [00:00<00:00, 5128.61it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.bias]  
Loading weights:  86%|████████▌ | 405/472 [00:00<00:00, 5127.11it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.bias]
Loading weights:  86%|████████▌ | 406/472 [00:00<00:00, 5137.16it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.weight]
Loading weights:  86%|████████▌ | 406/472 [00:00<00:00, 5135.80it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.weight]
Loading weights:  86%|████████▌ | 407/472 [00:00<00:00, 5145.80it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.bias]
Loading weights:  86%|████████▌ | 407/472 [00:00<00:00, 5144.23it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.bias]
Loading weights:  86%|████████▋ | 408/472 [00:00<00:00, 5154.26it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.weight]
Loading weights:  86%|████████▋ | 408/472 [00:00<00:00, 5152.77it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.weight]
Loading weights:  87%|████████▋ | 409/472 [00:00<00:00, 5162.65it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.bias]         
Loading weights:  87%|████████▋ | 409/472 [00:00<00:00, 5161.22it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.bias]
Loading weights:  87%|████████▋ | 410/472 [00:00<00:00, 5171.35it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.weight]
Loading weights:  87%|████████▋ | 410/472 [00:00<00:00, 5169.92it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.weight]
Loading weights:  87%|████████▋ | 411/472 [00:00<00:00, 5179.99it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.bias]    
Loading weights:  87%|████████▋ | 411/472 [00:00<00:00, 5178.42it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.bias]
Loading weights:  87%|████████▋ | 412/472 [00:00<00:00, 5188.20it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.weight]
Loading weights:  87%|████████▋ | 412/472 [00:00<00:00, 5186.76it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.weight]
Loading weights:  88%|████████▊ | 413/472 [00:00<00:00, 5196.87it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.bias]  
Loading weights:  88%|████████▊ | 413/472 [00:00<00:00, 5195.44it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.bias]
Loading weights:  88%|████████▊ | 414/472 [00:00<00:00, 5205.72it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.weight]
Loading weights:  88%|████████▊ | 414/472 [00:00<00:00, 5204.33it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.weight]
Loading weights:  88%|████████▊ | 415/472 [00:00<00:00, 5214.44it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.bias]      
Loading weights:  88%|████████▊ | 415/472 [00:00<00:00, 5212.98it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.bias]
Loading weights:  88%|████████▊ | 416/472 [00:00<00:00, 5222.98it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.weight]
Loading weights:  88%|████████▊ | 416/472 [00:00<00:00, 5221.56it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.weight]
Loading weights:  88%|████████▊ | 417/472 [00:00<00:00, 5231.73it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.bias]  
Loading weights:  88%|████████▊ | 417/472 [00:00<00:00, 5230.35it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.bias]
Loading weights:  89%|████████▊ | 418/472 [00:00<00:00, 5240.34it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.weight]
Loading weights:  89%|████████▊ | 418/472 [00:00<00:00, 5238.95it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.weight]
Loading weights:  89%|████████▉ | 419/472 [00:00<00:00, 5248.92it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.bias]
Loading weights:  89%|████████▉ | 419/472 [00:00<00:00, 5247.40it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.bias]
Loading weights:  89%|████████▉ | 420/472 [00:00<00:00, 5257.37it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.weight]
Loading weights:  89%|████████▉ | 420/472 [00:00<00:00, 5255.78it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.weight]
Loading weights:  89%|████████▉ | 421/472 [00:00<00:00, 5265.94it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.bias]         
Loading weights:  89%|████████▉ | 421/472 [00:00<00:00, 5264.42it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.bias]
Loading weights:  89%|████████▉ | 422/472 [00:00<00:00, 5274.55it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.weight]
Loading weights:  89%|████████▉ | 422/472 [00:00<00:00, 5273.09it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.weight]
Loading weights:  90%|████████▉ | 423/472 [00:00<00:00, 5282.56it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.bias]    
Loading weights:  90%|████████▉ | 423/472 [00:00<00:00, 5281.16it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.bias]
Loading weights:  90%|████████▉ | 424/472 [00:00<00:00, 5291.08it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.weight]
Loading weights:  90%|████████▉ | 424/472 [00:00<00:00, 5289.62it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.weight]
Loading weights:  90%|█████████ | 425/472 [00:00<00:00, 5299.59it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.bias]  
Loading weights:  90%|█████████ | 425/472 [00:00<00:00, 5298.20it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.bias]
Loading weights:  90%|█████████ | 426/472 [00:00<00:00, 5307.97it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.weight]
Loading weights:  90%|█████████ | 426/472 [00:00<00:00, 5306.58it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.weight]
Loading weights:  90%|█████████ | 427/472 [00:00<00:00, 5316.70it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.bias]      
Loading weights:  90%|█████████ | 427/472 [00:00<00:00, 5315.31it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.bias]
Loading weights:  91%|█████████ | 428/472 [00:00<00:00, 5325.23it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.weight]
Loading weights:  91%|█████████ | 428/472 [00:00<00:00, 5323.84it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.weight]
Loading weights:  91%|█████████ | 429/472 [00:00<00:00, 5333.97it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.bias]  
Loading weights:  91%|█████████ | 429/472 [00:00<00:00, 5332.58it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.bias]
Loading weights:  91%|█████████ | 430/472 [00:00<00:00, 5342.48it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.weight]
Loading weights:  91%|█████████ | 430/472 [00:00<00:00, 5341.02it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.weight]
Loading weights:  91%|█████████▏| 431/472 [00:00<00:00, 5351.11it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.bias]
Loading weights:  91%|█████████▏| 431/472 [00:00<00:00, 5349.53it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.bias]
Loading weights:  92%|█████████▏| 432/472 [00:00<00:00, 5359.34it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.weight]
Loading weights:  92%|█████████▏| 432/472 [00:00<00:00, 5357.82it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.weight]
Loading weights:  92%|█████████▏| 433/472 [00:00<00:00, 5367.82it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.bias]         
Loading weights:  92%|█████████▏| 433/472 [00:00<00:00, 5366.29it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.bias]
Loading weights:  92%|█████████▏| 434/472 [00:00<00:00, 5376.22it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.weight]
Loading weights:  92%|█████████▏| 434/472 [00:00<00:00, 5374.76it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.weight]
Loading weights:  92%|█████████▏| 435/472 [00:00<00:00, 5382.40it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.bias]    
Loading weights:  92%|█████████▏| 435/472 [00:00<00:00, 5380.01it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.bias]
Loading weights:  92%|█████████▏| 436/472 [00:00<00:00, 5389.17it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.weight]
Loading weights:  92%|█████████▏| 436/472 [00:00<00:00, 5387.50it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.weight]
Loading weights:  93%|█████████▎| 437/472 [00:00<00:00, 5397.41it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.bias]  
Loading weights:  93%|█████████▎| 437/472 [00:00<00:00, 5395.26it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.bias]
Loading weights:  93%|█████████▎| 438/472 [00:00<00:00, 5404.35it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.weight]
Loading weights:  93%|█████████▎| 438/472 [00:00<00:00, 5402.61it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.weight]
Loading weights:  93%|█████████▎| 439/472 [00:00<00:00, 5412.36it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.bias]      
Loading weights:  93%|█████████▎| 439/472 [00:00<00:00, 5410.81it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.bias]
Loading weights:  93%|█████████▎| 440/472 [00:00<00:00, 5419.92it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.weight]
Loading weights:  93%|█████████▎| 440/472 [00:00<00:00, 5418.38it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.weight]
Loading weights:  93%|█████████▎| 441/472 [00:00<00:00, 5428.09it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.bias]  
Loading weights:  93%|█████████▎| 441/472 [00:00<00:00, 5425.75it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.bias]
Loading weights:  94%|█████████▎| 442/472 [00:00<00:00, 5434.58it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.weight]
Loading weights:  94%|█████████▎| 442/472 [00:00<00:00, 5432.78it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.weight]
Loading weights:  94%|█████████▍| 443/472 [00:00<00:00, 5442.27it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.bias]
Loading weights:  94%|█████████▍| 443/472 [00:00<00:00, 5440.07it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.bias]
Loading weights:  94%|█████████▍| 444/472 [00:00<00:00, 5449.67it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.weight]
Loading weights:  94%|█████████▍| 444/472 [00:00<00:00, 5448.14it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.weight]
Loading weights:  94%|█████████▍| 445/472 [00:00<00:00, 5457.92it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.bias]         
Loading weights:  94%|█████████▍| 445/472 [00:00<00:00, 5456.37it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.bias]
Loading weights:  94%|█████████▍| 446/472 [00:00<00:00, 5466.23it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.weight]
Loading weights:  94%|█████████▍| 446/472 [00:00<00:00, 5464.68it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.weight]
Loading weights:  95%|█████████▍| 447/472 [00:00<00:00, 5474.52it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.bias]   
Loading weights:  95%|█████████▍| 447/472 [00:00<00:00, 5472.97it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.bias]
Loading weights:  95%|█████████▍| 448/472 [00:00<00:00, 5482.61it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.weight]
Loading weights:  95%|█████████▍| 448/472 [00:00<00:00, 5480.99it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.weight]
Loading weights:  95%|█████████▌| 449/472 [00:00<00:00, 5490.76it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.bias]  
Loading weights:  95%|█████████▌| 449/472 [00:00<00:00, 5489.27it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.bias]
Loading weights:  95%|█████████▌| 450/472 [00:00<00:00, 5499.01it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.weight]
Loading weights:  95%|█████████▌| 450/472 [00:00<00:00, 5497.41it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.weight]
Loading weights:  96%|█████████▌| 451/472 [00:00<00:00, 5507.06it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.bias]      
Loading weights:  96%|█████████▌| 451/472 [00:00<00:00, 5505.51it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.bias]
Loading weights:  96%|█████████▌| 452/472 [00:00<00:00, 5515.30it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.weight]
Loading weights:  96%|█████████▌| 452/472 [00:00<00:00, 5513.81it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.weight]
Loading weights:  96%|█████████▌| 453/472 [00:00<00:00, 5523.65it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.bias]  
Loading weights:  96%|█████████▌| 453/472 [00:00<00:00, 5522.17it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.bias]
Loading weights:  96%|█████████▌| 454/472 [00:00<00:00, 5531.94it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.weight]
Loading weights:  96%|█████████▌| 454/472 [00:00<00:00, 5530.44it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.weight]
Loading weights:  96%|█████████▋| 455/472 [00:00<00:00, 5540.15it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.bias]
Loading weights:  96%|█████████▋| 455/472 [00:00<00:00, 5538.65it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.bias]
Loading weights:  97%|█████████▋| 456/472 [00:00<00:00, 5548.34it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.weight]
Loading weights:  97%|█████████▋| 456/472 [00:00<00:00, 5546.72it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.weight]
Loading weights:  97%|█████████▋| 457/472 [00:00<00:00, 5556.43it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.bias]         
Loading weights:  97%|█████████▋| 457/472 [00:00<00:00, 5554.95it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.bias]
Loading weights:  97%|█████████▋| 458/472 [00:00<00:00, 5564.74it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.weight]
Loading weights:  97%|█████████▋| 458/472 [00:00<00:00, 5563.19it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.weight]
Loading weights:  97%|█████████▋| 459/472 [00:00<00:00, 5573.03it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.bias]    
Loading weights:  97%|█████████▋| 459/472 [00:00<00:00, 5571.61it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.bias]
Loading weights:  97%|█████████▋| 460/472 [00:00<00:00, 5581.16it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.weight]
Loading weights:  97%|█████████▋| 460/472 [00:00<00:00, 5579.63it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.weight]
Loading weights:  98%|█████████▊| 461/472 [00:00<00:00, 5589.38it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.bias]  
Loading weights:  98%|█████████▊| 461/472 [00:00<00:00, 5587.75it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.bias]
Loading weights:  98%|█████████▊| 462/472 [00:00<00:00, 5597.49it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.weight]
Loading weights:  98%|█████████▊| 462/472 [00:00<00:00, 5595.93it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.weight]
Loading weights:  98%|█████████▊| 463/472 [00:00<00:00, 5605.82it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.bias]      
Loading weights:  98%|█████████▊| 463/472 [00:00<00:00, 5604.40it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.bias]
Loading weights:  98%|█████████▊| 464/472 [00:00<00:00, 5614.12it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.weight]
Loading weights:  98%|█████████▊| 464/472 [00:00<00:00, 5612.61it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.weight]
Loading weights:  99%|█████████▊| 465/472 [00:00<00:00, 5622.39it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.bias]  
Loading weights:  99%|█████████▊| 465/472 [00:00<00:00, 5620.90it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.bias]
Loading weights:  99%|█████████▊| 466/472 [00:00<00:00, 5630.67it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.weight]
Loading weights:  99%|█████████▊| 466/472 [00:00<00:00, 5629.24it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.weight]
Loading weights:  99%|█████████▉| 467/472 [00:00<00:00, 5639.00it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.bias]
Loading weights:  99%|█████████▉| 467/472 [00:00<00:00, 5637.52it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.bias]
Loading weights:  99%|█████████▉| 468/472 [00:00<00:00, 5646.99it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.weight]
Loading weights:  99%|█████████▉| 468/472 [00:00<00:00, 5645.42it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.weight]
Loading weights:  99%|█████████▉| 469/472 [00:00<00:00, 5655.04it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.bias]         
Loading weights:  99%|█████████▉| 469/472 [00:00<00:00, 5653.55it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.bias]
Loading weights: 100%|█████████▉| 470/472 [00:00<00:00, 5663.00it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.weight]
Loading weights: 100%|█████████▉| 470/472 [00:00<00:00, 5661.48it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.weight]
Loading weights: 100%|█████████▉| 471/472 [00:00<00:00, 5671.02it/s, Materializing param=vision_model.post_layernorm.bias]                   
Loading weights: 100%|█████████▉| 471/472 [00:00<00:00, 5669.52it/s, Materializing param=vision_model.post_layernorm.bias]
Loading weights: 100%|██████████| 472/472 [00:00<00:00, 5679.15it/s, Materializing param=vision_model.post_layernorm.weight]
Loading weights: 100%|██████████| 472/472 [00:00<00:00, 5677.72it/s, Materializing param=vision_model.post_layernorm.weight]
Loading weights: 100%|██████████| 472/472 [00:00<00:00, 5673.50it/s, Materializing param=vision_model.post_layernorm.weight]
+The tied weights mapping and config for this model specifies to tie text_decoder.bert.embeddings.word_embeddings.weight to text_decoder.cls.predictions.decoder.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
+BlipForConditionalGeneration LOAD REPORT from: Salesforce/blip-image-captioning-base
+Key                                       | Status     |  | 
+------------------------------------------+------------+--+-
+text_decoder.bert.embeddings.position_ids | UNEXPECTED |  | 
+
+Notes:
+- UNEXPECTED	:can be ignored when loading from different task/architecture; not ok if you expect identical arch.
+
Loading weights:   0%|          | 0/788 [00:00<?, ?it/s]
Loading weights:   0%|          | 1/788 [00:00<00:00, 59074.70it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.bias]
Loading weights:   0%|          | 1/788 [00:00<00:00, 2337.96it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.bias] 
Loading weights:   0%|          | 2/788 [00:00<00:00, 1934.20it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.weight]
Loading weights:   0%|          | 2/788 [00:00<00:00, 1782.91it/s, Materializing param=text_decoder.bert.embeddings.LayerNorm.weight]
Loading weights:   0%|          | 3/788 [00:00<00:00, 2286.97it/s, Materializing param=text_decoder.bert.embeddings.position_embeddings.weight]
Loading weights:   0%|          | 3/788 [00:00<00:00, 1889.33it/s, Materializing param=text_decoder.bert.embeddings.position_embeddings.weight]
Loading weights:   1%|          | 4/788 [00:00<00:00, 1985.23it/s, Materializing param=text_decoder.bert.embeddings.word_embeddings.weight]    
Loading weights:   1%|          | 4/788 [00:00<00:00, 1628.07it/s, Materializing param=text_decoder.bert.embeddings.word_embeddings.weight]
Loading weights:   1%|          | 5/788 [00:00<00:00, 1722.51it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:   1%|          | 5/788 [00:00<00:00, 1698.37it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:   1%|          | 6/788 [00:00<00:00, 1987.66it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:   1%|          | 6/788 [00:00<00:00, 1967.31it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:   1%|          | 7/788 [00:00<00:00, 2009.87it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.bias]      
Loading weights:   1%|          | 7/788 [00:00<00:00, 1725.04it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.bias]
Loading weights:   1%|          | 8/788 [00:00<00:00, 1677.22it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.weight]
Loading weights:   1%|          | 8/788 [00:00<00:00, 1660.45it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.output.dense.weight]
Loading weights:   1%|          | 9/788 [00:00<00:00, 1832.64it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.bias]      
Loading weights:   1%|          | 9/788 [00:00<00:00, 1681.98it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.bias]
Loading weights:   1%|▏         | 10/788 [00:00<00:00, 1664.21it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.weight]
Loading weights:   1%|▏         | 10/788 [00:00<00:00, 1585.57it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.key.weight]
Loading weights:   1%|▏         | 11/788 [00:00<00:00, 1646.53it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.bias]
Loading weights:   1%|▏         | 11/788 [00:00<00:00, 1633.82it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.bias]
Loading weights:   2%|▏         | 12/788 [00:00<00:00, 1579.83it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.weight]
Loading weights:   2%|▏         | 12/788 [00:00<00:00, 1476.39it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.query.weight]
Loading weights:   2%|▏         | 13/788 [00:00<00:00, 1577.49it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.bias]  
Loading weights:   2%|▏         | 13/788 [00:00<00:00, 1507.99it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.bias]
Loading weights:   2%|▏         | 14/788 [00:00<00:00, 1462.63it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.weight]
Loading weights:   2%|▏         | 14/788 [00:00<00:00, 1458.06it/s, Materializing param=text_decoder.bert.encoder.layer.0.attention.self.value.weight]
Loading weights:   2%|▏         | 15/788 [00:00<00:00, 1503.80it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:   2%|▏         | 15/788 [00:00<00:00, 1499.25it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:   2%|▏         | 16/788 [00:00<00:00, 1511.46it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:   2%|▏         | 16/788 [00:00<00:00, 1498.43it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:   2%|▏         | 17/788 [00:00<00:00, 1437.16it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.bias]      
Loading weights:   2%|▏         | 17/788 [00:00<00:00, 1419.87it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.bias]
Loading weights:   2%|▏         | 18/788 [00:00<00:00, 1492.07it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:   2%|▏         | 18/788 [00:00<00:00, 1448.36it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:   2%|▏         | 19/788 [00:00<00:00, 1513.70it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.bias]      
Loading weights:   2%|▏         | 19/788 [00:00<00:00, 1501.41it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.bias]
Loading weights:   3%|▎         | 20/788 [00:00<00:00, 1523.15it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.weight]
Loading weights:   3%|▎         | 20/788 [00:00<00:00, 1511.73it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.key.weight]
Loading weights:   3%|▎         | 21/788 [00:00<00:00, 1484.21it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.bias]
Loading weights:   3%|▎         | 21/788 [00:00<00:00, 1478.48it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.bias]
Loading weights:   3%|▎         | 22/788 [00:00<00:00, 1536.55it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.weight]
Loading weights:   3%|▎         | 22/788 [00:00<00:00, 1529.93it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.query.weight]
Loading weights:   3%|▎         | 23/788 [00:00<00:00, 1537.87it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.bias]  
Loading weights:   3%|▎         | 23/788 [00:00<00:00, 1531.01it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.bias]
Loading weights:   3%|▎         | 24/788 [00:00<00:00, 1554.62it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.weight]
Loading weights:   3%|▎         | 24/788 [00:00<00:00, 1550.89it/s, Materializing param=text_decoder.bert.encoder.layer.0.crossattention.self.value.weight]
Loading weights:   3%|▎         | 25/788 [00:00<00:00, 1577.80it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.bias]         
Loading weights:   3%|▎         | 25/788 [00:00<00:00, 1567.21it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.bias]
Loading weights:   3%|▎         | 26/788 [00:00<00:00, 1542.59it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.weight]
Loading weights:   3%|▎         | 26/788 [00:00<00:00, 1520.12it/s, Materializing param=text_decoder.bert.encoder.layer.0.intermediate.dense.weight]
Loading weights:   3%|▎         | 27/788 [00:00<00:00, 1547.04it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.bias]    
Loading weights:   3%|▎         | 27/788 [00:00<00:00, 1541.71it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.bias]
Loading weights:   4%|▎         | 28/788 [00:00<00:00, 1513.78it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.weight]
Loading weights:   4%|▎         | 28/788 [00:00<00:00, 1508.63it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.LayerNorm.weight]
Loading weights:   4%|▎         | 29/788 [00:00<00:00, 1533.03it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.bias]      
Loading weights:   4%|▎         | 29/788 [00:00<00:00, 1529.07it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.bias]
Loading weights:   4%|▍         | 30/788 [00:00<00:00, 1570.61it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.weight]
Loading weights:   4%|▍         | 30/788 [00:00<00:00, 1539.74it/s, Materializing param=text_decoder.bert.encoder.layer.0.output.dense.weight]
Loading weights:   4%|▍         | 31/788 [00:00<00:00, 1580.93it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:   4%|▍         | 31/788 [00:00<00:00, 1577.30it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:   4%|▍         | 32/788 [00:00<00:00, 1600.82it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:   4%|▍         | 32/788 [00:00<00:00, 1586.46it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:   4%|▍         | 33/788 [00:00<00:00, 1600.40it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.bias]      
Loading weights:   4%|▍         | 33/788 [00:00<00:00, 1556.38it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.bias]
Loading weights:   4%|▍         | 34/788 [00:00<00:00, 1552.38it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.weight]
Loading weights:   4%|▍         | 34/788 [00:00<00:00, 1546.61it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.output.dense.weight]
Loading weights:   4%|▍         | 35/788 [00:00<00:00, 1553.91it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.bias]      
Loading weights:   4%|▍         | 35/788 [00:00<00:00, 1539.76it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.bias]
Loading weights:   5%|▍         | 36/788 [00:00<00:00, 1577.58it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.weight]
Loading weights:   5%|▍         | 36/788 [00:00<00:00, 1536.64it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.key.weight]
Loading weights:   5%|▍         | 37/788 [00:00<00:00, 1568.01it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.bias]
Loading weights:   5%|▍         | 37/788 [00:00<00:00, 1564.16it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.bias]
Loading weights:   5%|▍         | 38/788 [00:00<00:00, 1508.36it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.weight]
Loading weights:   5%|▍         | 38/788 [00:00<00:00, 1485.89it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.query.weight]
Loading weights:   5%|▍         | 39/788 [00:00<00:00, 1503.94it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.bias]  
Loading weights:   5%|▍         | 39/788 [00:00<00:00, 1501.68it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.bias]
Loading weights:   5%|▌         | 40/788 [00:00<00:00, 1494.00it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.weight]
Loading weights:   5%|▌         | 40/788 [00:00<00:00, 1485.22it/s, Materializing param=text_decoder.bert.encoder.layer.1.attention.self.value.weight]
Loading weights:   5%|▌         | 41/788 [00:00<00:00, 1500.03it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:   5%|▌         | 41/788 [00:00<00:00, 1497.44it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:   5%|▌         | 42/788 [00:00<00:00, 1497.33it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:   5%|▌         | 42/788 [00:00<00:00, 1468.90it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:   5%|▌         | 43/788 [00:00<00:00, 1427.05it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.bias]      
Loading weights:   5%|▌         | 43/788 [00:00<00:00, 1425.36it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.bias]
Loading weights:   6%|▌         | 44/788 [00:00<00:00, 1446.99it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:   6%|▌         | 44/788 [00:00<00:00, 1444.24it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:   6%|▌         | 45/788 [00:00<00:00, 1455.57it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.bias]      
Loading weights:   6%|▌         | 45/788 [00:00<00:00, 1440.97it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.bias]
Loading weights:   6%|▌         | 46/788 [00:00<00:00, 1454.14it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.weight]
Loading weights:   6%|▌         | 46/788 [00:00<00:00, 1452.44it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.key.weight]
Loading weights:   6%|▌         | 47/788 [00:00<00:00, 1469.81it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.bias]
Loading weights:   6%|▌         | 47/788 [00:00<00:00, 1467.71it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.bias]
Loading weights:   6%|▌         | 48/788 [00:00<00:00, 1477.80it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.weight]
Loading weights:   6%|▌         | 48/788 [00:00<00:00, 1464.18it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.query.weight]
Loading weights:   6%|▌         | 49/788 [00:00<00:00, 1482.82it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.bias]  
Loading weights:   6%|▌         | 49/788 [00:00<00:00, 1471.43it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.bias]
Loading weights:   6%|▋         | 50/788 [00:00<00:00, 1471.25it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.weight]
Loading weights:   6%|▋         | 50/788 [00:00<00:00, 1468.31it/s, Materializing param=text_decoder.bert.encoder.layer.1.crossattention.self.value.weight]
Loading weights:   6%|▋         | 51/788 [00:00<00:00, 1490.97it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.bias]         
Loading weights:   6%|▋         | 51/788 [00:00<00:00, 1462.62it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.bias]
Loading weights:   7%|▋         | 52/788 [00:00<00:00, 1457.74it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.weight]
Loading weights:   7%|▋         | 52/788 [00:00<00:00, 1440.65it/s, Materializing param=text_decoder.bert.encoder.layer.1.intermediate.dense.weight]
Loading weights:   7%|▋         | 53/788 [00:00<00:00, 1462.39it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.bias]    
Loading weights:   7%|▋         | 53/788 [00:00<00:00, 1450.82it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.bias]
Loading weights:   7%|▋         | 54/788 [00:00<00:00, 1459.98it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.weight]
Loading weights:   7%|▋         | 54/788 [00:00<00:00, 1446.68it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.LayerNorm.weight]
Loading weights:   7%|▋         | 55/788 [00:00<00:00, 1469.38it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.bias]      
Loading weights:   7%|▋         | 55/788 [00:00<00:00, 1457.07it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.bias]
Loading weights:   7%|▋         | 56/788 [00:00<00:00, 1467.25it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.weight]
Loading weights:   7%|▋         | 56/788 [00:00<00:00, 1447.60it/s, Materializing param=text_decoder.bert.encoder.layer.1.output.dense.weight]
Loading weights:   7%|▋         | 57/788 [00:00<00:00, 1440.67it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:   7%|▋         | 57/788 [00:00<00:00, 1439.48it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:   7%|▋         | 58/788 [00:00<00:00, 1461.37it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:   7%|▋         | 58/788 [00:00<00:00, 1440.85it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:   7%|▋         | 59/788 [00:00<00:00, 1461.99it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.bias]      
Loading weights:   7%|▋         | 59/788 [00:00<00:00, 1450.13it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.bias]
Loading weights:   8%|▊         | 60/788 [00:00<00:00, 1471.38it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.weight]
Loading weights:   8%|▊         | 60/788 [00:00<00:00, 1459.15it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.output.dense.weight]
Loading weights:   8%|▊         | 61/788 [00:00<00:00, 1472.97it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.bias]      
Loading weights:   8%|▊         | 61/788 [00:00<00:00, 1470.94it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.bias]
Loading weights:   8%|▊         | 62/788 [00:00<00:00, 1480.39it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.weight]
Loading weights:   8%|▊         | 62/788 [00:00<00:00, 1478.94it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.key.weight]
Loading weights:   8%|▊         | 63/788 [00:00<00:00, 1499.65it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.bias]
Loading weights:   8%|▊         | 63/788 [00:00<00:00, 1482.78it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.bias]
Loading weights:   8%|▊         | 64/788 [00:00<00:00, 1492.78it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.weight]
Loading weights:   8%|▊         | 64/788 [00:00<00:00, 1479.57it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.query.weight]
Loading weights:   8%|▊         | 65/788 [00:00<00:00, 1491.03it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.bias]  
Loading weights:   8%|▊         | 65/788 [00:00<00:00, 1466.67it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.bias]
Loading weights:   8%|▊         | 66/788 [00:00<00:00, 1467.00it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.weight]
Loading weights:   8%|▊         | 66/788 [00:00<00:00, 1450.65it/s, Materializing param=text_decoder.bert.encoder.layer.2.attention.self.value.weight]
Loading weights:   9%|▊         | 67/788 [00:00<00:00, 1458.65it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:   9%|▊         | 67/788 [00:00<00:00, 1449.19it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:   9%|▊         | 68/788 [00:00<00:00, 1459.11it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:   9%|▊         | 68/788 [00:00<00:00, 1458.05it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:   9%|▉         | 69/788 [00:00<00:00, 1477.40it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.bias]      
Loading weights:   9%|▉         | 69/788 [00:00<00:00, 1467.25it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.bias]
Loading weights:   9%|▉         | 70/788 [00:00<00:00, 1485.86it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:   9%|▉         | 70/788 [00:00<00:00, 1469.78it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:   9%|▉         | 71/788 [00:00<00:00, 1466.34it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.bias]      
Loading weights:   9%|▉         | 71/788 [00:00<00:00, 1457.25it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.bias]
Loading weights:   9%|▉         | 72/788 [00:00<00:00, 1464.10it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.weight]
Loading weights:   9%|▉         | 72/788 [00:00<00:00, 1454.49it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.key.weight]
Loading weights:   9%|▉         | 73/788 [00:00<00:00, 1455.41it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.bias]
Loading weights:   9%|▉         | 73/788 [00:00<00:00, 1454.54it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.bias]
Loading weights:   9%|▉         | 74/788 [00:00<00:00, 1462.95it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.weight]
Loading weights:   9%|▉         | 74/788 [00:00<00:00, 1461.65it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.query.weight]
Loading weights:  10%|▉         | 75/788 [00:00<00:00, 1472.76it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.bias]  
Loading weights:  10%|▉         | 75/788 [00:00<00:00, 1464.79it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.bias]
Loading weights:  10%|▉         | 76/788 [00:00<00:00, 1466.36it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  10%|▉         | 76/788 [00:00<00:00, 1458.85it/s, Materializing param=text_decoder.bert.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  10%|▉         | 77/788 [00:00<00:00, 1465.28it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.bias]         
Loading weights:  10%|▉         | 77/788 [00:00<00:00, 1462.10it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.bias]
Loading weights:  10%|▉         | 78/788 [00:00<00:00, 1453.24it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.weight]
Loading weights:  10%|▉         | 78/788 [00:00<00:00, 1451.35it/s, Materializing param=text_decoder.bert.encoder.layer.2.intermediate.dense.weight]
Loading weights:  10%|█         | 79/788 [00:00<00:00, 1458.70it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.bias]    
Loading weights:  10%|█         | 79/788 [00:00<00:00, 1445.43it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.bias]
Loading weights:  10%|█         | 80/788 [00:00<00:00, 1461.43it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  10%|█         | 80/788 [00:00<00:00, 1460.42it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  10%|█         | 81/788 [00:00<00:00, 1476.33it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.bias]      
Loading weights:  10%|█         | 81/788 [00:00<00:00, 1470.14it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.bias]
Loading weights:  10%|█         | 82/788 [00:00<00:00, 1456.31it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.weight]
Loading weights:  10%|█         | 82/788 [00:00<00:00, 1442.83it/s, Materializing param=text_decoder.bert.encoder.layer.2.output.dense.weight]
Loading weights:  11%|█         | 83/788 [00:00<00:00, 1458.27it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  11%|█         | 83/788 [00:00<00:00, 1455.33it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  11%|█         | 84/788 [00:00<00:00, 1458.13it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  11%|█         | 84/788 [00:00<00:00, 1451.08it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  11%|█         | 85/788 [00:00<00:00, 1466.50it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.bias]      
Loading weights:  11%|█         | 85/788 [00:00<00:00, 1454.66it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.bias]
Loading weights:  11%|█         | 86/788 [00:00<00:00, 1468.61it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.weight]
Loading weights:  11%|█         | 86/788 [00:00<00:00, 1467.21it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.output.dense.weight]
Loading weights:  11%|█         | 87/788 [00:00<00:00, 1476.28it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.bias]      
Loading weights:  11%|█         | 87/788 [00:00<00:00, 1469.34it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.bias]
Loading weights:  11%|█         | 88/788 [00:00<00:00, 1478.45it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.weight]
Loading weights:  11%|█         | 88/788 [00:00<00:00, 1472.51it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.key.weight]
Loading weights:  11%|█▏        | 89/788 [00:00<00:00, 1482.00it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.bias]
Loading weights:  11%|█▏        | 89/788 [00:00<00:00, 1473.83it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.bias]
Loading weights:  11%|█▏        | 90/788 [00:00<00:00, 1478.88it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.weight]
Loading weights:  11%|█▏        | 90/788 [00:00<00:00, 1477.30it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.query.weight]
Loading weights:  12%|█▏        | 91/788 [00:00<00:00, 1483.90it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.bias]  
Loading weights:  12%|█▏        | 91/788 [00:00<00:00, 1477.87it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.bias]
Loading weights:  12%|█▏        | 92/788 [00:00<00:00, 1491.76it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.weight]
Loading weights:  12%|█▏        | 92/788 [00:00<00:00, 1484.57it/s, Materializing param=text_decoder.bert.encoder.layer.3.attention.self.value.weight]
Loading weights:  12%|█▏        | 93/788 [00:00<00:00, 1490.06it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  12%|█▏        | 93/788 [00:00<00:00, 1488.84it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  12%|█▏        | 94/788 [00:00<00:00, 1501.43it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  12%|█▏        | 94/788 [00:00<00:00, 1472.69it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  12%|█▏        | 95/788 [00:00<00:00, 1482.73it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.bias]      
Loading weights:  12%|█▏        | 95/788 [00:00<00:00, 1481.18it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.bias]
Loading weights:  12%|█▏        | 96/788 [00:00<00:00, 1489.55it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  12%|█▏        | 96/788 [00:00<00:00, 1483.36it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  12%|█▏        | 97/788 [00:00<00:00, 1491.97it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.bias]      
Loading weights:  12%|█▏        | 97/788 [00:00<00:00, 1482.05it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.bias]
Loading weights:  12%|█▏        | 98/788 [00:00<00:00, 1484.97it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  12%|█▏        | 98/788 [00:00<00:00, 1477.87it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  13%|█▎        | 99/788 [00:00<00:00, 1491.46it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  13%|█▎        | 99/788 [00:00<00:00, 1489.20it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  13%|█▎        | 100/788 [00:00<00:00, 1478.95it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  13%|█▎        | 100/788 [00:00<00:00, 1472.43it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  13%|█▎        | 101/788 [00:00<00:00, 1479.12it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.bias]  
Loading weights:  13%|█▎        | 101/788 [00:00<00:00, 1477.69it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.bias]
Loading weights:  13%|█▎        | 102/788 [00:00<00:00, 1468.45it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  13%|█▎        | 102/788 [00:00<00:00, 1462.20it/s, Materializing param=text_decoder.bert.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  13%|█▎        | 103/788 [00:00<00:00, 1461.49it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.bias]         
Loading weights:  13%|█▎        | 103/788 [00:00<00:00, 1453.02it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.bias]
Loading weights:  13%|█▎        | 104/788 [00:00<00:00, 1456.42it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.weight]
Loading weights:  13%|█▎        | 104/788 [00:00<00:00, 1449.64it/s, Materializing param=text_decoder.bert.encoder.layer.3.intermediate.dense.weight]
Loading weights:  13%|█▎        | 105/788 [00:00<00:00, 1455.02it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.bias]    
Loading weights:  13%|█▎        | 105/788 [00:00<00:00, 1442.99it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.bias]
Loading weights:  13%|█▎        | 106/788 [00:00<00:00, 1448.86it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  13%|█▎        | 106/788 [00:00<00:00, 1435.39it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  14%|█▎        | 107/788 [00:00<00:00, 1430.56it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.bias]      
Loading weights:  14%|█▎        | 107/788 [00:00<00:00, 1422.18it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.bias]
Loading weights:  14%|█▎        | 108/788 [00:00<00:00, 1420.14it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.weight]
Loading weights:  14%|█▎        | 108/788 [00:00<00:00, 1419.46it/s, Materializing param=text_decoder.bert.encoder.layer.3.output.dense.weight]
Loading weights:  14%|█▍        | 109/788 [00:00<00:00, 1409.71it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  14%|█▍        | 109/788 [00:00<00:00, 1403.61it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  14%|█▍        | 110/788 [00:00<00:00, 1405.41it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  14%|█▍        | 110/788 [00:00<00:00, 1403.93it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  14%|█▍        | 111/788 [00:00<00:00, 1410.47it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.bias]      
Loading weights:  14%|█▍        | 111/788 [00:00<00:00, 1405.94it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.bias]
Loading weights:  14%|█▍        | 112/788 [00:00<00:00, 1404.80it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.weight]
Loading weights:  14%|█▍        | 112/788 [00:00<00:00, 1398.08it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.output.dense.weight]
Loading weights:  14%|█▍        | 113/788 [00:00<00:00, 1403.07it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.bias]      
Loading weights:  14%|█▍        | 113/788 [00:00<00:00, 1402.42it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.bias]
Loading weights:  14%|█▍        | 114/788 [00:00<00:00, 1408.07it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.weight]
Loading weights:  14%|█▍        | 114/788 [00:00<00:00, 1393.64it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.key.weight]
Loading weights:  15%|█▍        | 115/788 [00:00<00:00, 1403.14it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.bias]
Loading weights:  15%|█▍        | 115/788 [00:00<00:00, 1402.27it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.bias]
Loading weights:  15%|█▍        | 116/788 [00:00<00:00, 1412.95it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.weight]
Loading weights:  15%|█▍        | 116/788 [00:00<00:00, 1406.83it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.query.weight]
Loading weights:  15%|█▍        | 117/788 [00:00<00:00, 1411.24it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.bias]  
Loading weights:  15%|█▍        | 117/788 [00:00<00:00, 1400.53it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.bias]
Loading weights:  15%|█▍        | 118/788 [00:00<00:00, 1405.21it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.weight]
Loading weights:  15%|█▍        | 118/788 [00:00<00:00, 1404.45it/s, Materializing param=text_decoder.bert.encoder.layer.4.attention.self.value.weight]
Loading weights:  15%|█▌        | 119/788 [00:00<00:00, 1410.09it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  15%|█▌        | 119/788 [00:00<00:00, 1405.09it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  15%|█▌        | 120/788 [00:00<00:00, 1410.47it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  15%|█▌        | 120/788 [00:00<00:00, 1409.10it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  15%|█▌        | 121/788 [00:00<00:00, 1417.64it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.bias]      
Loading weights:  15%|█▌        | 121/788 [00:00<00:00, 1417.07it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.bias]
Loading weights:  15%|█▌        | 122/788 [00:00<00:00, 1422.86it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  15%|█▌        | 122/788 [00:00<00:00, 1421.40it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  16%|█▌        | 123/788 [00:00<00:00, 1423.81it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.bias]      
Loading weights:  16%|█▌        | 123/788 [00:00<00:00, 1418.33it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.bias]
Loading weights:  16%|█▌        | 124/788 [00:00<00:00, 1428.26it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  16%|█▌        | 124/788 [00:00<00:00, 1427.08it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  16%|█▌        | 125/788 [00:00<00:00, 1437.30it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  16%|█▌        | 125/788 [00:00<00:00, 1436.28it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  16%|█▌        | 126/788 [00:00<00:00, 1440.56it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  16%|█▌        | 126/788 [00:00<00:00, 1436.35it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  16%|█▌        | 127/788 [00:00<00:00, 1421.24it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.bias]  
Loading weights:  16%|█▌        | 127/788 [00:00<00:00, 1415.85it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.bias]
Loading weights:  16%|█▌        | 128/788 [00:00<00:00, 1425.58it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  16%|█▌        | 128/788 [00:00<00:00, 1419.04it/s, Materializing param=text_decoder.bert.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  16%|█▋        | 129/788 [00:00<00:00, 1415.16it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.bias]         
Loading weights:  16%|█▋        | 129/788 [00:00<00:00, 1414.59it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.bias]
Loading weights:  16%|█▋        | 130/788 [00:00<00:00, 1424.67it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.weight]
Loading weights:  16%|█▋        | 130/788 [00:00<00:00, 1423.99it/s, Materializing param=text_decoder.bert.encoder.layer.4.intermediate.dense.weight]
Loading weights:  17%|█▋        | 131/788 [00:00<00:00, 1433.75it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.bias]    
Loading weights:  17%|█▋        | 131/788 [00:00<00:00, 1433.10it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.bias]
Loading weights:  17%|█▋        | 132/788 [00:00<00:00, 1438.88it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 132/788 [00:00<00:00, 1438.35it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 133/788 [00:00<00:00, 1439.27it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.bias]      
Loading weights:  17%|█▋        | 133/788 [00:00<00:00, 1438.76it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.bias]
Loading weights:  17%|█▋        | 134/788 [00:00<00:00, 1443.05it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.weight]
Loading weights:  17%|█▋        | 134/788 [00:00<00:00, 1436.55it/s, Materializing param=text_decoder.bert.encoder.layer.4.output.dense.weight]
Loading weights:  17%|█▋        | 135/788 [00:00<00:00, 1446.00it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  17%|█▋        | 135/788 [00:00<00:00, 1440.65it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  17%|█▋        | 136/788 [00:00<00:00, 1440.89it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 136/788 [00:00<00:00, 1438.66it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  17%|█▋        | 137/788 [00:00<00:00, 1444.04it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.bias]      
Loading weights:  17%|█▋        | 137/788 [00:00<00:00, 1439.26it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.bias]
Loading weights:  18%|█▊        | 138/788 [00:00<00:00, 1443.83it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.weight]
Loading weights:  18%|█▊        | 138/788 [00:00<00:00, 1443.20it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.output.dense.weight]
Loading weights:  18%|█▊        | 139/788 [00:00<00:00, 1450.06it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.bias]      
Loading weights:  18%|█▊        | 139/788 [00:00<00:00, 1449.61it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.bias]
Loading weights:  18%|█▊        | 140/788 [00:00<00:00, 1445.54it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.weight]
Loading weights:  18%|█▊        | 140/788 [00:00<00:00, 1440.66it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.key.weight]
Loading weights:  18%|█▊        | 141/788 [00:00<00:00, 1448.59it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.bias]
Loading weights:  18%|█▊        | 141/788 [00:00<00:00, 1443.89it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.bias]
Loading weights:  18%|█▊        | 142/788 [00:00<00:00, 1443.21it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.weight]
Loading weights:  18%|█▊        | 142/788 [00:00<00:00, 1442.77it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.query.weight]
Loading weights:  18%|█▊        | 143/788 [00:00<00:00, 1442.78it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.bias]  
Loading weights:  18%|█▊        | 143/788 [00:00<00:00, 1438.42it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.bias]
Loading weights:  18%|█▊        | 144/788 [00:00<00:00, 1447.28it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.weight]
Loading weights:  18%|█▊        | 144/788 [00:00<00:00, 1440.87it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.weight]
Loading weights:  18%|█▊        | 145/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.attention.self.value.weight]
Loading weights:  18%|█▊        | 145/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  18%|█▊        | 145/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  19%|█▊        | 146/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  19%|█▊        | 146/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  19%|█▊        | 147/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.bias]      
Loading weights:  19%|█▊        | 147/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.bias]
Loading weights:  19%|█▉        | 148/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  19%|█▉        | 148/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  19%|█▉        | 149/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.bias]      
Loading weights:  19%|█▉        | 149/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.bias]
Loading weights:  19%|█▉        | 150/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  19%|█▉        | 150/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  19%|█▉        | 151/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  19%|█▉        | 151/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  19%|█▉        | 152/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  19%|█▉        | 152/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  19%|█▉        | 153/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.bias]  
Loading weights:  19%|█▉        | 153/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.bias]
Loading weights:  20%|█▉        | 154/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  20%|█▉        | 154/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  20%|█▉        | 155/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.bias]         
Loading weights:  20%|█▉        | 155/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.bias]
Loading weights:  20%|█▉        | 156/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.weight]
Loading weights:  20%|█▉        | 156/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.intermediate.dense.weight]
Loading weights:  20%|█▉        | 157/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.bias]    
Loading weights:  20%|█▉        | 157/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.bias]
Loading weights:  20%|██        | 158/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  20%|██        | 158/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  20%|██        | 159/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.bias]      
Loading weights:  20%|██        | 159/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.bias]
Loading weights:  20%|██        | 160/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.weight]
Loading weights:  20%|██        | 160/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.5.output.dense.weight]
Loading weights:  20%|██        | 161/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  20%|██        | 161/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  21%|██        | 162/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  21%|██        | 162/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  21%|██        | 163/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.bias]      
Loading weights:  21%|██        | 163/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.bias]
Loading weights:  21%|██        | 164/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.weight]
Loading weights:  21%|██        | 164/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.output.dense.weight]
Loading weights:  21%|██        | 165/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.bias]      
Loading weights:  21%|██        | 165/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.bias]
Loading weights:  21%|██        | 166/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.weight]
Loading weights:  21%|██        | 166/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.key.weight]
Loading weights:  21%|██        | 167/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.bias]
Loading weights:  21%|██        | 167/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.bias]
Loading weights:  21%|██▏       | 168/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.weight]
Loading weights:  21%|██▏       | 168/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.query.weight]
Loading weights:  21%|██▏       | 169/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.bias]  
Loading weights:  21%|██▏       | 169/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.bias]
Loading weights:  22%|██▏       | 170/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.weight]
Loading weights:  22%|██▏       | 170/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.attention.self.value.weight]
Loading weights:  22%|██▏       | 171/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  22%|██▏       | 171/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  22%|██▏       | 172/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  22%|██▏       | 172/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  22%|██▏       | 173/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.bias]      
Loading weights:  22%|██▏       | 173/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.bias]
Loading weights:  22%|██▏       | 174/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  22%|██▏       | 174/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  22%|██▏       | 175/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.bias]      
Loading weights:  22%|██▏       | 175/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.bias]
Loading weights:  22%|██▏       | 176/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  22%|██▏       | 176/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  22%|██▏       | 177/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  22%|██▏       | 177/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  23%|██▎       | 178/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  23%|██▎       | 178/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  23%|██▎       | 179/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.bias]  
Loading weights:  23%|██▎       | 179/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.bias]
Loading weights:  23%|██▎       | 180/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  23%|██▎       | 180/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  23%|██▎       | 181/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.bias]         
Loading weights:  23%|██▎       | 181/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.bias]
Loading weights:  23%|██▎       | 182/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.weight]
Loading weights:  23%|██▎       | 182/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.intermediate.dense.weight]
Loading weights:  23%|██▎       | 183/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.bias]    
Loading weights:  23%|██▎       | 183/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.bias]
Loading weights:  23%|██▎       | 184/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  23%|██▎       | 184/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  23%|██▎       | 185/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.bias]      
Loading weights:  23%|██▎       | 185/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.bias]
Loading weights:  24%|██▎       | 186/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.weight]
Loading weights:  24%|██▎       | 186/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.6.output.dense.weight]
Loading weights:  24%|██▎       | 187/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  24%|██▎       | 187/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  24%|██▍       | 188/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  24%|██▍       | 188/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  24%|██▍       | 189/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.bias]      
Loading weights:  24%|██▍       | 189/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.bias]
Loading weights:  24%|██▍       | 190/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.weight]
Loading weights:  24%|██▍       | 190/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.output.dense.weight]
Loading weights:  24%|██▍       | 191/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.bias]      
Loading weights:  24%|██▍       | 191/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.bias]
Loading weights:  24%|██▍       | 192/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.weight]
Loading weights:  24%|██▍       | 192/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.key.weight]
Loading weights:  24%|██▍       | 193/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.bias]
Loading weights:  24%|██▍       | 193/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.bias]
Loading weights:  25%|██▍       | 194/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.weight]
Loading weights:  25%|██▍       | 194/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.query.weight]
Loading weights:  25%|██▍       | 195/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.bias]  
Loading weights:  25%|██▍       | 195/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.bias]
Loading weights:  25%|██▍       | 196/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.weight]
Loading weights:  25%|██▍       | 196/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.attention.self.value.weight]
Loading weights:  25%|██▌       | 197/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  25%|██▌       | 197/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  25%|██▌       | 198/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  25%|██▌       | 198/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  25%|██▌       | 199/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.bias]      
Loading weights:  25%|██▌       | 199/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.bias]
Loading weights:  25%|██▌       | 200/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  25%|██▌       | 200/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  26%|██▌       | 201/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.bias]      
Loading weights:  26%|██▌       | 201/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.bias]
Loading weights:  26%|██▌       | 202/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  26%|██▌       | 202/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  26%|██▌       | 203/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  26%|██▌       | 203/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  26%|██▌       | 204/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  26%|██▌       | 204/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  26%|██▌       | 205/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.bias]  
Loading weights:  26%|██▌       | 205/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.bias]
Loading weights:  26%|██▌       | 206/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  26%|██▌       | 206/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  26%|██▋       | 207/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.bias]         
Loading weights:  26%|██▋       | 207/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.bias]
Loading weights:  26%|██▋       | 208/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.weight]
Loading weights:  26%|██▋       | 208/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.intermediate.dense.weight]
Loading weights:  27%|██▋       | 209/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.bias]    
Loading weights:  27%|██▋       | 209/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.bias]
Loading weights:  27%|██▋       | 210/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  27%|██▋       | 210/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  27%|██▋       | 211/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.bias]      
Loading weights:  27%|██▋       | 211/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.bias]
Loading weights:  27%|██▋       | 212/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.weight]
Loading weights:  27%|██▋       | 212/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.7.output.dense.weight]
Loading weights:  27%|██▋       | 213/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  27%|██▋       | 213/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  27%|██▋       | 214/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  27%|██▋       | 214/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  27%|██▋       | 215/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.bias]      
Loading weights:  27%|██▋       | 215/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.bias]
Loading weights:  27%|██▋       | 216/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.weight]
Loading weights:  27%|██▋       | 216/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.output.dense.weight]
Loading weights:  28%|██▊       | 217/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.bias]      
Loading weights:  28%|██▊       | 217/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.bias]
Loading weights:  28%|██▊       | 218/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.weight]
Loading weights:  28%|██▊       | 218/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.key.weight]
Loading weights:  28%|██▊       | 219/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.bias]
Loading weights:  28%|██▊       | 219/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.bias]
Loading weights:  28%|██▊       | 220/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.weight]
Loading weights:  28%|██▊       | 220/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.query.weight]
Loading weights:  28%|██▊       | 221/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.bias]  
Loading weights:  28%|██▊       | 221/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.bias]
Loading weights:  28%|██▊       | 222/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.weight]
Loading weights:  28%|██▊       | 222/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.attention.self.value.weight]
Loading weights:  28%|██▊       | 223/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  28%|██▊       | 223/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  28%|██▊       | 224/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  28%|██▊       | 224/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  29%|██▊       | 225/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.bias]      
Loading weights:  29%|██▊       | 225/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.bias]
Loading weights:  29%|██▊       | 226/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  29%|██▊       | 226/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  29%|██▉       | 227/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.bias]      
Loading weights:  29%|██▉       | 227/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.bias]
Loading weights:  29%|██▉       | 228/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  29%|██▉       | 228/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  29%|██▉       | 229/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  29%|██▉       | 229/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  29%|██▉       | 230/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  29%|██▉       | 230/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  29%|██▉       | 231/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.bias]  
Loading weights:  29%|██▉       | 231/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.bias]
Loading weights:  29%|██▉       | 232/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  29%|██▉       | 232/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  30%|██▉       | 233/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.bias]         
Loading weights:  30%|██▉       | 233/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.bias]
Loading weights:  30%|██▉       | 234/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.weight]
Loading weights:  30%|██▉       | 234/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.intermediate.dense.weight]
Loading weights:  30%|██▉       | 235/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.bias]    
Loading weights:  30%|██▉       | 235/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.bias]
Loading weights:  30%|██▉       | 236/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  30%|██▉       | 236/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  30%|███       | 237/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.bias]      
Loading weights:  30%|███       | 237/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.bias]
Loading weights:  30%|███       | 238/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.weight]
Loading weights:  30%|███       | 238/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.8.output.dense.weight]
Loading weights:  30%|███       | 239/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  30%|███       | 239/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  30%|███       | 240/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  30%|███       | 240/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  31%|███       | 241/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.bias]      
Loading weights:  31%|███       | 241/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.bias]
Loading weights:  31%|███       | 242/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.weight]
Loading weights:  31%|███       | 242/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.output.dense.weight]
Loading weights:  31%|███       | 243/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.bias]      
Loading weights:  31%|███       | 243/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.bias]
Loading weights:  31%|███       | 244/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.weight]
Loading weights:  31%|███       | 244/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.key.weight]
Loading weights:  31%|███       | 245/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.bias]
Loading weights:  31%|███       | 245/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.bias]
Loading weights:  31%|███       | 246/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.weight]
Loading weights:  31%|███       | 246/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.query.weight]
Loading weights:  31%|███▏      | 247/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.bias]  
Loading weights:  31%|███▏      | 247/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.bias]
Loading weights:  31%|███▏      | 248/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.weight]
Loading weights:  31%|███▏      | 248/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.attention.self.value.weight]
Loading weights:  32%|███▏      | 249/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  32%|███▏      | 249/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  32%|███▏      | 250/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  32%|███▏      | 250/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  32%|███▏      | 251/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.bias]      
Loading weights:  32%|███▏      | 251/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.bias]
Loading weights:  32%|███▏      | 252/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  32%|███▏      | 252/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  32%|███▏      | 253/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.bias]      
Loading weights:  32%|███▏      | 253/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.bias]
Loading weights:  32%|███▏      | 254/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  32%|███▏      | 254/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  32%|███▏      | 255/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  32%|███▏      | 255/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  32%|███▏      | 256/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  32%|███▏      | 256/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  33%|███▎      | 257/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.bias]  
Loading weights:  33%|███▎      | 257/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.bias]
Loading weights:  33%|███▎      | 258/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  33%|███▎      | 258/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  33%|███▎      | 259/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.bias]         
Loading weights:  33%|███▎      | 259/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.bias]
Loading weights:  33%|███▎      | 260/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.weight]
Loading weights:  33%|███▎      | 260/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.intermediate.dense.weight]
Loading weights:  33%|███▎      | 261/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.bias]    
Loading weights:  33%|███▎      | 261/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.bias]
Loading weights:  33%|███▎      | 262/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  33%|███▎      | 262/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  33%|███▎      | 263/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.bias]      
Loading weights:  33%|███▎      | 263/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.bias]
Loading weights:  34%|███▎      | 264/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.weight]
Loading weights:  34%|███▎      | 264/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.9.output.dense.weight]
Loading weights:  34%|███▎      | 265/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  34%|███▎      | 265/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  34%|███▍      | 266/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  34%|███▍      | 266/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  34%|███▍      | 267/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.bias]      
Loading weights:  34%|███▍      | 267/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.bias]
Loading weights:  34%|███▍      | 268/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.weight]
Loading weights:  34%|███▍      | 268/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.output.dense.weight]
Loading weights:  34%|███▍      | 269/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.bias]      
Loading weights:  34%|███▍      | 269/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.bias]
Loading weights:  34%|███▍      | 270/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.weight]
Loading weights:  34%|███▍      | 270/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.key.weight]
Loading weights:  34%|███▍      | 271/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.bias]
Loading weights:  34%|███▍      | 271/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.bias]
Loading weights:  35%|███▍      | 272/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.weight]
Loading weights:  35%|███▍      | 272/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.query.weight]
Loading weights:  35%|███▍      | 273/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.bias]  
Loading weights:  35%|███▍      | 273/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.bias]
Loading weights:  35%|███▍      | 274/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.weight]
Loading weights:  35%|███▍      | 274/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.attention.self.value.weight]
Loading weights:  35%|███▍      | 275/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  35%|███▍      | 275/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  35%|███▌      | 276/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  35%|███▌      | 276/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  35%|███▌      | 277/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.bias]      
Loading weights:  35%|███▌      | 277/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.bias]
Loading weights:  35%|███▌      | 278/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  35%|███▌      | 278/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  35%|███▌      | 279/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.bias]      
Loading weights:  35%|███▌      | 279/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.bias]
Loading weights:  36%|███▌      | 280/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  36%|███▌      | 280/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  36%|███▌      | 281/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  36%|███▌      | 281/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  36%|███▌      | 282/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  36%|███▌      | 282/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  36%|███▌      | 283/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.bias]  
Loading weights:  36%|███▌      | 283/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.bias]
Loading weights:  36%|███▌      | 284/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  36%|███▌      | 284/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  36%|███▌      | 285/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.bias]         
Loading weights:  36%|███▌      | 285/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.bias]
Loading weights:  36%|███▋      | 286/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.weight]
Loading weights:  36%|███▋      | 286/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.intermediate.dense.weight]
Loading weights:  36%|███▋      | 287/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.bias]    
Loading weights:  36%|███▋      | 287/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.bias]
Loading weights:  37%|███▋      | 288/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  37%|███▋      | 288/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  37%|███▋      | 289/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.bias]      
Loading weights:  37%|███▋      | 289/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.bias]
Loading weights:  37%|███▋      | 290/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.weight]
Loading weights:  37%|███▋      | 290/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.10.output.dense.weight]
Loading weights:  37%|███▋      | 291/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  37%|███▋      | 291/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  37%|███▋      | 292/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  37%|███▋      | 292/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  37%|███▋      | 293/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.bias]      
Loading weights:  37%|███▋      | 293/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.bias]
Loading weights:  37%|███▋      | 294/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.weight]
Loading weights:  37%|███▋      | 294/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.output.dense.weight]
Loading weights:  37%|███▋      | 295/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.bias]      
Loading weights:  37%|███▋      | 295/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.bias]
Loading weights:  38%|███▊      | 296/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.weight]
Loading weights:  38%|███▊      | 296/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.key.weight]
Loading weights:  38%|███▊      | 297/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.bias]
Loading weights:  38%|███▊      | 297/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.bias]
Loading weights:  38%|███▊      | 298/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.weight]
Loading weights:  38%|███▊      | 298/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.query.weight]
Loading weights:  38%|███▊      | 299/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.bias]  
Loading weights:  38%|███▊      | 299/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.bias]
Loading weights:  38%|███▊      | 300/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.weight]
Loading weights:  38%|███▊      | 300/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.attention.self.value.weight]
Loading weights:  38%|███▊      | 301/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  38%|███▊      | 301/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  38%|███▊      | 302/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  38%|███▊      | 302/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  38%|███▊      | 303/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.bias]      
Loading weights:  38%|███▊      | 303/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.bias]
Loading weights:  39%|███▊      | 304/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  39%|███▊      | 304/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  39%|███▊      | 305/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.bias]      
Loading weights:  39%|███▊      | 305/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.bias]
Loading weights:  39%|███▉      | 306/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  39%|███▉      | 306/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  39%|███▉      | 307/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  39%|███▉      | 307/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  39%|███▉      | 308/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  39%|███▉      | 308/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  39%|███▉      | 309/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.bias]  
Loading weights:  39%|███▉      | 309/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.bias]
Loading weights:  39%|███▉      | 310/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  39%|███▉      | 310/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  39%|███▉      | 311/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.bias]         
Loading weights:  39%|███▉      | 311/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.bias]
Loading weights:  40%|███▉      | 312/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.weight]
Loading weights:  40%|███▉      | 312/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.intermediate.dense.weight]
Loading weights:  40%|███▉      | 313/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.bias]    
Loading weights:  40%|███▉      | 313/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.bias]
Loading weights:  40%|███▉      | 314/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  40%|███▉      | 314/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  40%|███▉      | 315/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.bias]      
Loading weights:  40%|███▉      | 315/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.bias]
Loading weights:  40%|████      | 316/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.weight]
Loading weights:  40%|████      | 316/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.bert.encoder.layer.11.output.dense.weight]
Loading weights:  40%|████      | 317/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.bias]                     
Loading weights:  40%|████      | 317/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.bias]
Loading weights:  40%|████      | 318/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.decoder.weight]
Loading weights:  40%|████      | 318/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.decoder.weight]
Loading weights:  40%|████      | 319/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.bias]
Loading weights:  40%|████      | 319/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.bias]
Loading weights:  41%|████      | 320/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.weight]
Loading weights:  41%|████      | 320/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.LayerNorm.weight]
Loading weights:  41%|████      | 321/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.dense.bias]      
Loading weights:  41%|████      | 321/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.dense.bias]
Loading weights:  41%|████      | 322/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.dense.weight]
Loading weights:  41%|████      | 322/788 [00:00<00:00, 1449.57it/s, Materializing param=text_decoder.cls.predictions.transform.dense.weight]
Loading weights:  41%|████      | 323/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.LayerNorm.bias]             
Loading weights:  41%|████      | 323/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.LayerNorm.bias]
Loading weights:  41%|████      | 324/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.LayerNorm.weight]
Loading weights:  41%|████      | 324/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.LayerNorm.weight]
Loading weights:  41%|████      | 325/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.position_embeddings.weight]
Loading weights:  41%|████      | 325/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.position_embeddings.weight]
Loading weights:  41%|████▏     | 326/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.word_embeddings.weight]    
Loading weights:  41%|████▏     | 326/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.embeddings.word_embeddings.weight]
Loading weights:  41%|████▏     | 327/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:  41%|████▏     | 327/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.LayerNorm.bias]
Loading weights:  42%|████▏     | 328/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:  42%|████▏     | 328/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.LayerNorm.weight]
Loading weights:  42%|████▏     | 329/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.dense.bias]      
Loading weights:  42%|████▏     | 329/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.dense.bias]
Loading weights:  42%|████▏     | 330/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.dense.weight]
Loading weights:  42%|████▏     | 330/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.output.dense.weight]
Loading weights:  42%|████▏     | 331/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.key.bias]      
Loading weights:  42%|████▏     | 331/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.key.bias]
Loading weights:  42%|████▏     | 332/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.key.weight]
Loading weights:  42%|████▏     | 332/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.key.weight]
Loading weights:  42%|████▏     | 333/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.query.bias]
Loading weights:  42%|████▏     | 333/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.query.bias]
Loading weights:  42%|████▏     | 334/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.query.weight]
Loading weights:  42%|████▏     | 334/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.query.weight]
Loading weights:  43%|████▎     | 335/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.value.bias]  
Loading weights:  43%|████▎     | 335/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.value.bias]
Loading weights:  43%|████▎     | 336/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.value.weight]
Loading weights:  43%|████▎     | 336/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.attention.self.value.weight]
Loading weights:  43%|████▎     | 337/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:  43%|████▎     | 337/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.LayerNorm.bias]
Loading weights:  43%|████▎     | 338/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:  43%|████▎     | 338/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.LayerNorm.weight]
Loading weights:  43%|████▎     | 339/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.dense.bias]      
Loading weights:  43%|████▎     | 339/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.dense.bias]
Loading weights:  43%|████▎     | 340/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:  43%|████▎     | 340/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.output.dense.weight]
Loading weights:  43%|████▎     | 341/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.key.bias]      
Loading weights:  43%|████▎     | 341/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.key.bias]
Loading weights:  43%|████▎     | 342/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.key.weight]
Loading weights:  43%|████▎     | 342/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.key.weight]
Loading weights:  44%|████▎     | 343/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.query.bias]
Loading weights:  44%|████▎     | 343/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.query.bias]
Loading weights:  44%|████▎     | 344/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.query.weight]
Loading weights:  44%|████▎     | 344/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.query.weight]
Loading weights:  44%|████▍     | 345/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.value.bias]  
Loading weights:  44%|████▍     | 345/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.value.bias]
Loading weights:  44%|████▍     | 346/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.value.weight]
Loading weights:  44%|████▍     | 346/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.crossattention.self.value.weight]
Loading weights:  44%|████▍     | 347/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.intermediate.dense.bias]         
Loading weights:  44%|████▍     | 347/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.intermediate.dense.bias]
Loading weights:  44%|████▍     | 348/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.intermediate.dense.weight]
Loading weights:  44%|████▍     | 348/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.intermediate.dense.weight]
Loading weights:  44%|████▍     | 349/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.LayerNorm.bias]    
Loading weights:  44%|████▍     | 349/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.LayerNorm.bias]
Loading weights:  44%|████▍     | 350/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.LayerNorm.weight]
Loading weights:  44%|████▍     | 350/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.LayerNorm.weight]
Loading weights:  45%|████▍     | 351/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.dense.bias]      
Loading weights:  45%|████▍     | 351/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.dense.bias]
Loading weights:  45%|████▍     | 352/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.dense.weight]
Loading weights:  45%|████▍     | 352/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.0.output.dense.weight]
Loading weights:  45%|████▍     | 353/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:  45%|████▍     | 353/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.LayerNorm.bias]
Loading weights:  45%|████▍     | 354/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:  45%|████▍     | 354/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.LayerNorm.weight]
Loading weights:  45%|████▌     | 355/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.dense.bias]      
Loading weights:  45%|████▌     | 355/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.dense.bias]
Loading weights:  45%|████▌     | 356/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.dense.weight]
Loading weights:  45%|████▌     | 356/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.output.dense.weight]
Loading weights:  45%|████▌     | 357/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.key.bias]      
Loading weights:  45%|████▌     | 357/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.key.bias]
Loading weights:  45%|████▌     | 358/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.key.weight]
Loading weights:  45%|████▌     | 358/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.key.weight]
Loading weights:  46%|████▌     | 359/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.query.bias]
Loading weights:  46%|████▌     | 359/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.query.bias]
Loading weights:  46%|████▌     | 360/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.query.weight]
Loading weights:  46%|████▌     | 360/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.query.weight]
Loading weights:  46%|████▌     | 361/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.value.bias]  
Loading weights:  46%|████▌     | 361/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.value.bias]
Loading weights:  46%|████��     | 362/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.value.weight]
Loading weights:  46%|████▌     | 362/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.attention.self.value.weight]
Loading weights:  46%|████▌     | 363/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:  46%|████▌     | 363/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.LayerNorm.bias]
Loading weights:  46%|████▌     | 364/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:  46%|████▌     | 364/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.LayerNorm.weight]
Loading weights:  46%|████▋     | 365/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.dense.bias]      
Loading weights:  46%|████▋     | 365/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.dense.bias]
Loading weights:  46%|████▋     | 366/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:  46%|████▋     | 366/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.output.dense.weight]
Loading weights:  47%|████▋     | 367/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.key.bias]      
Loading weights:  47%|████▋     | 367/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.key.bias]
Loading weights:  47%|████▋     | 368/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.key.weight]
Loading weights:  47%|████▋     | 368/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.key.weight]
Loading weights:  47%|████▋     | 369/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.query.bias]
Loading weights:  47%|████▋     | 369/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.query.bias]
Loading weights:  47%|████▋     | 370/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.query.weight]
Loading weights:  47%|████▋     | 370/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.query.weight]
Loading weights:  47%|████▋     | 371/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.value.bias]  
Loading weights:  47%|████▋     | 371/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.value.bias]
Loading weights:  47%|████▋     | 372/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.value.weight]
Loading weights:  47%|████▋     | 372/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.crossattention.self.value.weight]
Loading weights:  47%|████▋     | 373/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.intermediate.dense.bias]         
Loading weights:  47%|████▋     | 373/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.intermediate.dense.bias]
Loading weights:  47%|████▋     | 374/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.intermediate.dense.weight]
Loading weights:  47%|████▋     | 374/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.intermediate.dense.weight]
Loading weights:  48%|████▊     | 375/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.LayerNorm.bias]    
Loading weights:  48%|████▊     | 375/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.LayerNorm.bias]
Loading weights:  48%|████▊     | 376/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.LayerNorm.weight]
Loading weights:  48%|████▊     | 376/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.LayerNorm.weight]
Loading weights:  48%|████▊     | 377/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.dense.bias]      
Loading weights:  48%|████▊     | 377/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.dense.bias]
Loading weights:  48%|████▊     | 378/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.dense.weight]
Loading weights:  48%|████▊     | 378/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.1.output.dense.weight]
Loading weights:  48%|████▊     | 379/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:  48%|████▊     | 379/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.LayerNorm.bias]
Loading weights:  48%|████▊     | 380/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:  48%|████▊     | 380/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.LayerNorm.weight]
Loading weights:  48%|████▊     | 381/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.dense.bias]      
Loading weights:  48%|████▊     | 381/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.dense.bias]
Loading weights:  48%|████▊     | 382/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.dense.weight]
Loading weights:  48%|████▊     | 382/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.output.dense.weight]
Loading weights:  49%|████▊     | 383/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.key.bias]      
Loading weights:  49%|████▊     | 383/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.key.bias]
Loading weights:  49%|████▊     | 384/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.key.weight]
Loading weights:  49%|████▊     | 384/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.key.weight]
Loading weights:  49%|████▉     | 385/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.query.bias]
Loading weights:  49%|████▉     | 385/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.query.bias]
Loading weights:  49%|████▉     | 386/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.query.weight]
Loading weights:  49%|████▉     | 386/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.query.weight]
Loading weights:  49%|████▉     | 387/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.value.bias]  
Loading weights:  49%|████▉     | 387/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.value.bias]
Loading weights:  49%|████▉     | 388/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.value.weight]
Loading weights:  49%|████▉     | 388/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.attention.self.value.weight]
Loading weights:  49%|████▉     | 389/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:  49%|████▉     | 389/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.LayerNorm.bias]
Loading weights:  49%|████▉     | 390/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:  49%|████▉     | 390/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.LayerNorm.weight]
Loading weights:  50%|████▉     | 391/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.dense.bias]      
Loading weights:  50%|████▉     | 391/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.dense.bias]
Loading weights:  50%|████▉     | 392/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:  50%|████▉     | 392/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.output.dense.weight]
Loading weights:  50%|████▉     | 393/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.key.bias]      
Loading weights:  50%|████▉     | 393/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.key.bias]
Loading weights:  50%|█████     | 394/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.key.weight]
Loading weights:  50%|█████     | 394/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.key.weight]
Loading weights:  50%|█████     | 395/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.query.bias]
Loading weights:  50%|█████     | 395/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.query.bias]
Loading weights:  50%|█████     | 396/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.query.weight]
Loading weights:  50%|█████     | 396/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.query.weight]
Loading weights:  50%|█████     | 397/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.value.bias]  
Loading weights:  50%|█████     | 397/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.value.bias]
Loading weights:  51%|█████     | 398/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  51%|█████     | 398/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.crossattention.self.value.weight]
Loading weights:  51%|█████     | 399/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.intermediate.dense.bias]         
Loading weights:  51%|█████     | 399/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.intermediate.dense.bias]
Loading weights:  51%|█████     | 400/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.intermediate.dense.weight]
Loading weights:  51%|█████     | 400/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.intermediate.dense.weight]
Loading weights:  51%|█████     | 401/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.LayerNorm.bias]    
Loading weights:  51%|█████     | 401/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.LayerNorm.bias]
Loading weights:  51%|█████     | 402/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  51%|█████     | 402/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.LayerNorm.weight]
Loading weights:  51%|█████     | 403/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.dense.bias]      
Loading weights:  51%|█████     | 403/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.dense.bias]
Loading weights:  51%|█████▏    | 404/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.dense.weight]
Loading weights:  51%|█████▏    | 404/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.2.output.dense.weight]
Loading weights:  51%|█████▏    | 405/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  51%|█████▏    | 405/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.LayerNorm.bias]
Loading weights:  52%|█████▏    | 406/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  52%|█████▏    | 406/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.LayerNorm.weight]
Loading weights:  52%|█████▏    | 407/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.dense.bias]      
Loading weights:  52%|█████▏    | 407/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.dense.bias]
Loading weights:  52%|█████▏    | 408/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.dense.weight]
Loading weights:  52%|█████▏    | 408/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.output.dense.weight]
Loading weights:  52%|█████▏    | 409/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.key.bias]      
Loading weights:  52%|█████▏    | 409/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.key.bias]
Loading weights:  52%|█████▏    | 410/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.key.weight]
Loading weights:  52%|█████▏    | 410/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.key.weight]
Loading weights:  52%|█████▏    | 411/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.query.bias]
Loading weights:  52%|█████▏    | 411/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.query.bias]
Loading weights:  52%|█████▏    | 412/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.query.weight]
Loading weights:  52%|█████▏    | 412/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.query.weight]
Loading weights:  52%|█████▏    | 413/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.value.bias]  
Loading weights:  52%|█████▏    | 413/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.value.bias]
Loading weights:  53%|█████▎    | 414/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.value.weight]
Loading weights:  53%|█████▎    | 414/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.attention.self.value.weight]
Loading weights:  53%|█████▎    | 415/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  53%|█████▎    | 415/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.LayerNorm.bias]
Loading weights:  53%|█████▎    | 416/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  53%|█████▎    | 416/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.LayerNorm.weight]
Loading weights:  53%|█████▎    | 417/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.dense.bias]      
Loading weights:  53%|█████▎    | 417/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.dense.bias]
Loading weights:  53%|█████▎    | 418/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  53%|█████▎    | 418/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.output.dense.weight]
Loading weights:  53%|█████▎    | 419/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.key.bias]      
Loading weights:  53%|█████▎    | 419/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.key.bias]
Loading weights:  53%|█████▎    | 420/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  53%|█████▎    | 420/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.key.weight]
Loading weights:  53%|█████▎    | 421/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  53%|█████▎    | 421/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.query.bias]
Loading weights:  54%|█████▎    | 422/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  54%|█████▎    | 422/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.query.weight]
Loading weights:  54%|█████▎    | 423/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.value.bias]  
Loading weights:  54%|█████▎    | 423/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.value.bias]
Loading weights:  54%|█████▍    | 424/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  54%|█████▍    | 424/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.crossattention.self.value.weight]
Loading weights:  54%|█████▍    | 425/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.intermediate.dense.bias]         
Loading weights:  54%|█████▍    | 425/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.intermediate.dense.bias]
Loading weights:  54%|█████��    | 426/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.intermediate.dense.weight]
Loading weights:  54%|█████▍    | 426/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.intermediate.dense.weight]
Loading weights:  54%|█████▍    | 427/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.LayerNorm.bias]    
Loading weights:  54%|█████▍    | 427/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.LayerNorm.bias]
Loading weights:  54%|█████▍    | 428/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  54%|█████▍    | 428/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.LayerNorm.weight]
Loading weights:  54%|█████▍    | 429/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.dense.bias]      
Loading weights:  54%|█████▍    | 429/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.dense.bias]
Loading weights:  55%|█████▍    | 430/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.dense.weight]
Loading weights:  55%|█████▍    | 430/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.3.output.dense.weight]
Loading weights:  55%|█████▍    | 431/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  55%|█████▍    | 431/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.LayerNorm.bias]
Loading weights:  55%|█████▍    | 432/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  55%|█████▍    | 432/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.LayerNorm.weight]
Loading weights:  55%|█████▍    | 433/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.dense.bias]      
Loading weights:  55%|█████▍    | 433/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.dense.bias]
Loading weights:  55%|█████▌    | 434/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.dense.weight]
Loading weights:  55%|█████▌    | 434/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.output.dense.weight]
Loading weights:  55%|█████▌    | 435/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.key.bias]      
Loading weights:  55%|█████▌    | 435/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.key.bias]
Loading weights:  55%|█████▌    | 436/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.key.weight]
Loading weights:  55%|█████▌    | 436/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.key.weight]
Loading weights:  55%|█████▌    | 437/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.query.bias]
Loading weights:  55%|█████▌    | 437/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.query.bias]
Loading weights:  56%|█████▌    | 438/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.query.weight]
Loading weights:  56%|█████▌    | 438/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.query.weight]
Loading weights:  56%|█████▌    | 439/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.value.bias]  
Loading weights:  56%|█████▌    | 439/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.value.bias]
Loading weights:  56%|█████▌    | 440/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.value.weight]
Loading weights:  56%|█████▌    | 440/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.attention.self.value.weight]
Loading weights:  56%|█████▌    | 441/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  56%|█████▌    | 441/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.LayerNorm.bias]
Loading weights:  56%|█████▌    | 442/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  56%|█████▌    | 442/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.LayerNorm.weight]
Loading weights:  56%|█████▌    | 443/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.dense.bias]      
Loading weights:  56%|█████▌    | 443/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.dense.bias]
Loading weights:  56%|█████▋    | 444/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  56%|█████▋    | 444/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.output.dense.weight]
Loading weights:  56%|█████▋    | 445/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.key.bias]      
Loading weights:  56%|█████▋    | 445/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.key.bias]
Loading weights:  57%|█████▋    | 446/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  57%|█████▋    | 446/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.key.weight]
Loading weights:  57%|█████▋    | 447/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  57%|█████▋    | 447/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.query.bias]
Loading weights:  57%|█████▋    | 448/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  57%|█████▋    | 448/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.query.weight]
Loading weights:  57%|█████▋    | 449/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.value.bias]  
Loading weights:  57%|█████▋    | 449/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.value.bias]
Loading weights:  57%|█████▋    | 450/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  57%|█████▋    | 450/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.crossattention.self.value.weight]
Loading weights:  57%|█████▋    | 451/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.intermediate.dense.bias]         
Loading weights:  57%|█████▋    | 451/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.intermediate.dense.bias]
Loading weights:  57%|█████▋    | 452/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.intermediate.dense.weight]
Loading weights:  57%|█████▋    | 452/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.intermediate.dense.weight]
Loading weights:  57%|█████▋    | 453/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.LayerNorm.bias]    
Loading weights:  57%|█████▋    | 453/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.LayerNorm.bias]
Loading weights:  58%|█████▊    | 454/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  58%|█████▊    | 454/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.LayerNorm.weight]
Loading weights:  58%|█████▊    | 455/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.dense.bias]      
Loading weights:  58%|█████▊    | 455/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.dense.bias]
Loading weights:  58%|█████▊    | 456/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.dense.weight]
Loading weights:  58%|█████▊    | 456/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.4.output.dense.weight]
Loading weights:  58%|█████▊    | 457/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  58%|█████▊    | 457/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.LayerNorm.bias]
Loading weights:  58%|█████▊    | 458/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  58%|█████▊    | 458/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.LayerNorm.weight]
Loading weights:  58%|█████▊    | 459/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.dense.bias]      
Loading weights:  58%|█████▊    | 459/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.dense.bias]
Loading weights:  58%|█████▊    | 460/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.dense.weight]
Loading weights:  58%|█████▊    | 460/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.output.dense.weight]
Loading weights:  59%|█████▊    | 461/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.key.bias]      
Loading weights:  59%|█████▊    | 461/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.key.bias]
Loading weights:  59%|█████▊    | 462/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.key.weight]
Loading weights:  59%|█████▊    | 462/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.key.weight]
Loading weights:  59%|█████▉    | 463/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.query.bias]
Loading weights:  59%|█████▉    | 463/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.query.bias]
Loading weights:  59%|█████▉    | 464/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.query.weight]
Loading weights:  59%|█████▉    | 464/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.query.weight]
Loading weights:  59%|█████▉    | 465/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.value.bias]  
Loading weights:  59%|█████▉    | 465/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.value.bias]
Loading weights:  59%|█████▉    | 466/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.value.weight]
Loading weights:  59%|█████▉    | 466/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.attention.self.value.weight]
Loading weights:  59%|█████▉    | 467/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  59%|█████▉    | 467/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.LayerNorm.bias]
Loading weights:  59%|█████▉    | 468/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  59%|█████▉    | 468/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.LayerNorm.weight]
Loading weights:  60%|█████▉    | 469/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.dense.bias]      
Loading weights:  60%|█████▉    | 469/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.dense.bias]
Loading weights:  60%|█████▉    | 470/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  60%|█████▉    | 470/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.output.dense.weight]
Loading weights:  60%|█████▉    | 471/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.key.bias]      
Loading weights:  60%|█████▉    | 471/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.key.bias]
Loading weights:  60%|█████▉    | 472/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  60%|█████▉    | 472/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.key.weight]
Loading weights:  60%|██████    | 473/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  60%|██████    | 473/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.query.bias]
Loading weights:  60%|██████    | 474/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  60%|██████    | 474/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.query.weight]
Loading weights:  60%|██████    | 475/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.value.bias]  
Loading weights:  60%|██████    | 475/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.value.bias]
Loading weights:  60%|██████    | 476/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  60%|██████    | 476/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.crossattention.self.value.weight]
Loading weights:  61%|██████    | 477/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.intermediate.dense.bias]         
Loading weights:  61%|██████    | 477/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.intermediate.dense.bias]
Loading weights:  61%|██████    | 478/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.intermediate.dense.weight]
Loading weights:  61%|██████    | 478/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.intermediate.dense.weight]
Loading weights:  61%|██████    | 479/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.LayerNorm.bias]    
Loading weights:  61%|██████    | 479/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.LayerNorm.bias]
Loading weights:  61%|██████    | 480/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  61%|██████    | 480/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.LayerNorm.weight]
Loading weights:  61%|██████    | 481/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.dense.bias]      
Loading weights:  61%|██████    | 481/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.dense.bias]
Loading weights:  61%|██████    | 482/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.dense.weight]
Loading weights:  61%|██████    | 482/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.5.output.dense.weight]
Loading weights:  61%|██████▏   | 483/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  61%|██████▏   | 483/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.LayerNorm.bias]
Loading weights:  61%|██████▏   | 484/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  61%|██████▏   | 484/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.LayerNorm.weight]
Loading weights:  62%|██████▏   | 485/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.dense.bias]      
Loading weights:  62%|██████▏   | 485/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.dense.bias]
Loading weights:  62%|██████▏   | 486/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.dense.weight]
Loading weights:  62%|██████▏   | 486/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.output.dense.weight]
Loading weights:  62%|██████▏   | 487/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.key.bias]      
Loading weights:  62%|██████▏   | 487/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.key.bias]
Loading weights:  62%|██████▏   | 488/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.key.weight]
Loading weights:  62%|██████▏   | 488/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.key.weight]
Loading weights:  62%|██████▏   | 489/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.query.bias]
Loading weights:  62%|██████▏   | 489/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.query.bias]
Loading weights:  62%|██████▏   | 490/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.query.weight]
Loading weights:  62%|██████▏   | 490/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.query.weight]
Loading weights:  62%|██████▏   | 491/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.value.bias]  
Loading weights:  62%|██████▏   | 491/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.value.bias]
Loading weights:  62%|██████▏   | 492/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.value.weight]
Loading weights:  62%|██████▏   | 492/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.attention.self.value.weight]
Loading weights:  63%|██████▎   | 493/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  63%|██████▎   | 493/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.LayerNorm.bias]
Loading weights:  63%|██████▎   | 494/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  63%|██████▎   | 494/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.LayerNorm.weight]
Loading weights:  63%|██████▎   | 495/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.dense.bias]      
Loading weights:  63%|██████▎   | 495/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.dense.bias]
Loading weights:  63%|██████▎   | 496/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  63%|██████▎   | 496/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.output.dense.weight]
Loading weights:  63%|██████▎   | 497/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.key.bias]      
Loading weights:  63%|██████▎   | 497/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.key.bias]
Loading weights:  63%|██████▎   | 498/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  63%|██████▎   | 498/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.key.weight]
Loading weights:  63%|██████▎   | 499/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  63%|██████▎   | 499/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.query.bias]
Loading weights:  63%|██████▎   | 500/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  63%|██████▎   | 500/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.query.weight]
Loading weights:  64%|██████▎   | 501/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.value.bias]  
Loading weights:  64%|██████▎   | 501/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.value.bias]
Loading weights:  64%|██████▎   | 502/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  64%|██████▎   | 502/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.crossattention.self.value.weight]
Loading weights:  64%|██████▍   | 503/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.intermediate.dense.bias]         
Loading weights:  64%|██████▍   | 503/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.intermediate.dense.bias]
Loading weights:  64%|██████▍   | 504/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.intermediate.dense.weight]
Loading weights:  64%|██████▍   | 504/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.intermediate.dense.weight]
Loading weights:  64%|██████▍   | 505/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.LayerNorm.bias]    
Loading weights:  64%|██████▍   | 505/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.LayerNorm.bias]
Loading weights:  64%|██████▍   | 506/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  64%|██████▍   | 506/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.LayerNorm.weight]
Loading weights:  64%|██████▍   | 507/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.dense.bias]      
Loading weights:  64%|██████▍   | 507/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.dense.bias]
Loading weights:  64%|██████▍   | 508/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.dense.weight]
Loading weights:  64%|██████▍   | 508/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.6.output.dense.weight]
Loading weights:  65%|██████▍   | 509/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  65%|██████▍   | 509/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.LayerNorm.bias]
Loading weights:  65%|██████▍   | 510/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  65%|██████▍   | 510/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.LayerNorm.weight]
Loading weights:  65%|██████▍   | 511/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.dense.bias]      
Loading weights:  65%|██████▍   | 511/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.dense.bias]
Loading weights:  65%|██████▍   | 512/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.dense.weight]
Loading weights:  65%|██████▍   | 512/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.output.dense.weight]
Loading weights:  65%|██████▌   | 513/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.key.bias]      
Loading weights:  65%|██████▌   | 513/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.key.bias]
Loading weights:  65%|██████▌   | 514/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.key.weight]
Loading weights:  65%|██████▌   | 514/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.key.weight]
Loading weights:  65%|██████▌   | 515/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.query.bias]
Loading weights:  65%|██████▌   | 515/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.query.bias]
Loading weights:  65%|██████▌   | 516/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.query.weight]
Loading weights:  65%|██████▌   | 516/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.query.weight]
Loading weights:  66%|██████▌   | 517/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.value.bias]  
Loading weights:  66%|██████▌   | 517/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.value.bias]
Loading weights:  66%|██████▌   | 518/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.value.weight]
Loading weights:  66%|██████▌   | 518/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.attention.self.value.weight]
Loading weights:  66%|██████▌   | 519/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  66%|██████▌   | 519/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.LayerNorm.bias]
Loading weights:  66%|██████▌   | 520/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  66%|██████▌   | 520/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.LayerNorm.weight]
Loading weights:  66%|██████▌   | 521/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.dense.bias]      
Loading weights:  66%|██████▌   | 521/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.dense.bias]
Loading weights:  66%|██████▌   | 522/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  66%|██████▌   | 522/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.output.dense.weight]
Loading weights:  66%|██████▋   | 523/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.key.bias]      
Loading weights:  66%|██████▋   | 523/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.key.bias]
Loading weights:  66%|██████▋   | 524/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  66%|██████▋   | 524/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.key.weight]
Loading weights:  67%|██████▋   | 525/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  67%|██████▋   | 525/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.query.bias]
Loading weights:  67%|██████▋   | 526/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  67%|██████▋   | 526/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.query.weight]
Loading weights:  67%|██████▋   | 527/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.value.bias]  
Loading weights:  67%|██████▋   | 527/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.value.bias]
Loading weights:  67%|██████▋   | 528/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  67%|██████▋   | 528/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.crossattention.self.value.weight]
Loading weights:  67%|██████▋   | 529/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.intermediate.dense.bias]         
Loading weights:  67%|██████▋   | 529/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.intermediate.dense.bias]
Loading weights:  67%|██████▋   | 530/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.intermediate.dense.weight]
Loading weights:  67%|██████▋   | 530/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.intermediate.dense.weight]
Loading weights:  67%|██████▋   | 531/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.LayerNorm.bias]    
Loading weights:  67%|██████▋   | 531/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.LayerNorm.bias]
Loading weights:  68%|██████▊   | 532/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  68%|██████▊   | 532/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.LayerNorm.weight]
Loading weights:  68%|██████▊   | 533/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.dense.bias]      
Loading weights:  68%|██████▊   | 533/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.dense.bias]
Loading weights:  68%|██████▊   | 534/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.dense.weight]
Loading weights:  68%|██████▊   | 534/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.7.output.dense.weight]
Loading weights:  68%|██████▊   | 535/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  68%|██████▊   | 535/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.LayerNorm.bias]
Loading weights:  68%|██████▊   | 536/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  68%|██████▊   | 536/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.LayerNorm.weight]
Loading weights:  68%|██████▊   | 537/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.dense.bias]      
Loading weights:  68%|██████▊   | 537/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.dense.bias]
Loading weights:  68%|██████▊   | 538/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.dense.weight]
Loading weights:  68%|██████▊   | 538/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.output.dense.weight]
Loading weights:  68%|██████▊   | 539/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.key.bias]      
Loading weights:  68%|██████▊   | 539/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.key.bias]
Loading weights:  69%|██████▊   | 540/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.key.weight]
Loading weights:  69%|██████▊   | 540/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.key.weight]
Loading weights:  69%|██████▊   | 541/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.query.bias]
Loading weights:  69%|██████▊   | 541/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.query.bias]
Loading weights:  69%|██████▉   | 542/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.query.weight]
Loading weights:  69%|██████▉   | 542/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.query.weight]
Loading weights:  69%|██████▉   | 543/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.value.bias]  
Loading weights:  69%|██████▉   | 543/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.value.bias]
Loading weights:  69%|██████▉   | 544/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.value.weight]
Loading weights:  69%|██████▉   | 544/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.attention.self.value.weight]
Loading weights:  69%|██████▉   | 545/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  69%|██████▉   | 545/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.LayerNorm.bias]
Loading weights:  69%|██████▉   | 546/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  69%|██████▉   | 546/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.LayerNorm.weight]
Loading weights:  69%|██████▉   | 547/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.dense.bias]      
Loading weights:  69%|██████▉   | 547/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.dense.bias]
Loading weights:  70%|██████▉   | 548/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  70%|██████▉   | 548/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.output.dense.weight]
Loading weights:  70%|██████▉   | 549/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.key.bias]      
Loading weights:  70%|██████▉   | 549/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.key.bias]
Loading weights:  70%|██████▉   | 550/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  70%|██████▉   | 550/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.key.weight]
Loading weights:  70%|██████▉   | 551/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  70%|██████▉   | 551/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.query.bias]
Loading weights:  70%|███████   | 552/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  70%|███████   | 552/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.query.weight]
Loading weights:  70%|███████   | 553/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.value.bias]  
Loading weights:  70%|███████   | 553/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.value.bias]
Loading weights:  70%|███████   | 554/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  70%|███████   | 554/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.crossattention.self.value.weight]
Loading weights:  70%|███████   | 555/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.intermediate.dense.bias]         
Loading weights:  70%|███████   | 555/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.intermediate.dense.bias]
Loading weights:  71%|███████   | 556/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.intermediate.dense.weight]
Loading weights:  71%|███████   | 556/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.intermediate.dense.weight]
Loading weights:  71%|███████   | 557/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.LayerNorm.bias]    
Loading weights:  71%|███████   | 557/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.LayerNorm.bias]
Loading weights:  71%|███████   | 558/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  71%|███████   | 558/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.LayerNorm.weight]
Loading weights:  71%|███████   | 559/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.dense.bias]      
Loading weights:  71%|███████   | 559/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.dense.bias]
Loading weights:  71%|███████   | 560/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.dense.weight]
Loading weights:  71%|███████   | 560/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.8.output.dense.weight]
Loading weights:  71%|███████   | 561/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  71%|███████   | 561/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.LayerNorm.bias]
Loading weights:  71%|███████▏  | 562/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  71%|███████▏  | 562/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.LayerNorm.weight]
Loading weights:  71%|███████▏  | 563/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.dense.bias]      
Loading weights:  71%|███████▏  | 563/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.dense.bias]
Loading weights:  72%|███████▏  | 564/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.dense.weight]
Loading weights:  72%|███████▏  | 564/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.output.dense.weight]
Loading weights:  72%|███████▏  | 565/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.key.bias]      
Loading weights:  72%|███████▏  | 565/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.key.bias]
Loading weights:  72%|███████▏  | 566/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.key.weight]
Loading weights:  72%|███████▏  | 566/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.key.weight]
Loading weights:  72%|███████▏  | 567/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.query.bias]
Loading weights:  72%|███████▏  | 567/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.query.bias]
Loading weights:  72%|███████▏  | 568/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.query.weight]
Loading weights:  72%|███████▏  | 568/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.query.weight]
Loading weights:  72%|███████▏  | 569/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.value.bias]  
Loading weights:  72%|███████▏  | 569/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.value.bias]
Loading weights:  72%|███████▏  | 570/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.value.weight]
Loading weights:  72%|███████▏  | 570/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.attention.self.value.weight]
Loading weights:  72%|███████▏  | 571/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  72%|███████▏  | 571/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.LayerNorm.bias]
Loading weights:  73%|███████▎  | 572/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  73%|███████▎  | 572/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.LayerNorm.weight]
Loading weights:  73%|███████▎  | 573/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.dense.bias]      
Loading weights:  73%|███████▎  | 573/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.dense.bias]
Loading weights:  73%|███████▎  | 574/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  73%|███████▎  | 574/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.output.dense.weight]
Loading weights:  73%|███████▎  | 575/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.key.bias]      
Loading weights:  73%|███████▎  | 575/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.key.bias]
Loading weights:  73%|███████▎  | 576/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  73%|███████▎  | 576/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.key.weight]
Loading weights:  73%|███████▎  | 577/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  73%|███████▎  | 577/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.query.bias]
Loading weights:  73%|███████▎  | 578/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  73%|███████▎  | 578/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.query.weight]
Loading weights:  73%|███████▎  | 579/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.value.bias]  
Loading weights:  73%|███████▎  | 579/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.value.bias]
Loading weights:  74%|███████▎  | 580/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  74%|███████▎  | 580/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.crossattention.self.value.weight]
Loading weights:  74%|███████▎  | 581/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.intermediate.dense.bias]         
Loading weights:  74%|███████▎  | 581/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.intermediate.dense.bias]
Loading weights:  74%|███████▍  | 582/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.intermediate.dense.weight]
Loading weights:  74%|███████▍  | 582/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.intermediate.dense.weight]
Loading weights:  74%|███████▍  | 583/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.LayerNorm.bias]    
Loading weights:  74%|███████▍  | 583/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.LayerNorm.bias]
Loading weights:  74%|███████▍  | 584/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  74%|███████▍  | 584/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.LayerNorm.weight]
Loading weights:  74%|███████▍  | 585/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.dense.bias]      
Loading weights:  74%|███████▍  | 585/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.dense.bias]
Loading weights:  74%|███████▍  | 586/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.dense.weight]
Loading weights:  74%|███████▍  | 586/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.9.output.dense.weight]
Loading weights:  74%|███████▍  | 587/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  74%|███████▍  | 587/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.LayerNorm.bias]
Loading weights:  75%|███████▍  | 588/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  75%|███████▍  | 588/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.LayerNorm.weight]
Loading weights:  75%|███████▍  | 589/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.dense.bias]      
Loading weights:  75%|███████▍  | 589/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.dense.bias]
Loading weights:  75%|███████▍  | 590/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.dense.weight]
Loading weights:  75%|███████▍  | 590/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.output.dense.weight]
Loading weights:  75%|███████▌  | 591/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.key.bias]      
Loading weights:  75%|███████▌  | 591/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.key.bias]
Loading weights:  75%|███████▌  | 592/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.key.weight]
Loading weights:  75%|███████▌  | 592/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.key.weight]
Loading weights:  75%|███████▌  | 593/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.query.bias]
Loading weights:  75%|███████▌  | 593/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.query.bias]
Loading weights:  75%|███████▌  | 594/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.query.weight]
Loading weights:  75%|███████▌  | 594/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.query.weight]
Loading weights:  76%|███████▌  | 595/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.value.bias]  
Loading weights:  76%|███████▌  | 595/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.value.bias]
Loading weights:  76%|███████▌  | 596/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.value.weight]
Loading weights:  76%|███████▌  | 596/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.attention.self.value.weight]
Loading weights:  76%|███████▌  | 597/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  76%|███████▌  | 597/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.LayerNorm.bias]
Loading weights:  76%|███████▌  | 598/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  76%|███████▌  | 598/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.LayerNorm.weight]
Loading weights:  76%|███████▌  | 599/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.dense.bias]      
Loading weights:  76%|███████▌  | 599/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.dense.bias]
Loading weights:  76%|███████▌  | 600/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  76%|███████▌  | 600/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.output.dense.weight]
Loading weights:  76%|███████▋  | 601/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.key.bias]      
Loading weights:  76%|███████▋  | 601/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.key.bias]
Loading weights:  76%|███████▋  | 602/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  76%|███████▋  | 602/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.key.weight]
Loading weights:  77%|███████▋  | 603/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  77%|███████▋  | 603/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.query.bias]
Loading weights:  77%|███████▋  | 604/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  77%|███████▋  | 604/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.query.weight]
Loading weights:  77%|███████▋  | 605/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.value.bias]  
Loading weights:  77%|███████▋  | 605/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.value.bias]
Loading weights:  77%|███████▋  | 606/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  77%|███████▋  | 606/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.crossattention.self.value.weight]
Loading weights:  77%|███████▋  | 607/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.intermediate.dense.bias]         
Loading weights:  77%|███████▋  | 607/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.intermediate.dense.bias]
Loading weights:  77%|███████▋  | 608/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.intermediate.dense.weight]
Loading weights:  77%|███████▋  | 608/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.intermediate.dense.weight]
Loading weights:  77%|███████▋  | 609/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.LayerNorm.bias]    
Loading weights:  77%|███████▋  | 609/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.LayerNorm.bias]
Loading weights:  77%|███████▋  | 610/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  77%|███████▋  | 610/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.LayerNorm.weight]
Loading weights:  78%|███████▊  | 611/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.dense.bias]      
Loading weights:  78%|███████▊  | 611/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.dense.bias]
Loading weights:  78%|███████▊  | 612/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.dense.weight]
Loading weights:  78%|███████▊  | 612/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.10.output.dense.weight]
Loading weights:  78%|███████▊  | 613/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  78%|███████▊  | 613/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.LayerNorm.bias]
Loading weights:  78%|███████▊  | 614/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  78%|███████▊  | 614/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.LayerNorm.weight]
Loading weights:  78%|███████▊  | 615/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.dense.bias]      
Loading weights:  78%|███████▊  | 615/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.dense.bias]
Loading weights:  78%|███████▊  | 616/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.dense.weight]
Loading weights:  78%|███████▊  | 616/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.output.dense.weight]
Loading weights:  78%|███████▊  | 617/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.key.bias]      
Loading weights:  78%|███████▊  | 617/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.key.bias]
Loading weights:  78%|███████▊  | 618/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.key.weight]
Loading weights:  78%|███████▊  | 618/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.key.weight]
Loading weights:  79%|███████▊  | 619/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.query.bias]
Loading weights:  79%|███████▊  | 619/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.query.bias]
Loading weights:  79%|███████▊  | 620/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.query.weight]
Loading weights:  79%|███████▊  | 620/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.query.weight]
Loading weights:  79%|███████▉  | 621/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.value.bias]  
Loading weights:  79%|███████▉  | 621/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.value.bias]
Loading weights:  79%|███████▉  | 622/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.value.weight]
Loading weights:  79%|███████▉  | 622/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.attention.self.value.weight]
Loading weights:  79%|███████▉  | 623/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  79%|███████▉  | 623/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.LayerNorm.bias]
Loading weights:  79%|███████▉  | 624/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  79%|███████▉  | 624/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.LayerNorm.weight]
Loading weights:  79%|███████▉  | 625/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.dense.bias]      
Loading weights:  79%|███████▉  | 625/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.dense.bias]
Loading weights:  79%|███████▉  | 626/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  79%|███████▉  | 626/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.output.dense.weight]
Loading weights:  80%|███████▉  | 627/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.key.bias]      
Loading weights:  80%|███████▉  | 627/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.key.bias]
Loading weights:  80%|███████▉  | 628/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  80%|███████▉  | 628/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.key.weight]
Loading weights:  80%|███████▉  | 629/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  80%|███████▉  | 629/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.query.bias]
Loading weights:  80%|███████▉  | 630/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  80%|███████▉  | 630/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.query.weight]
Loading weights:  80%|████████  | 631/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.value.bias]  
Loading weights:  80%|████████  | 631/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.value.bias]
Loading weights:  80%|████████  | 632/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  80%|████████  | 632/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.crossattention.self.value.weight]
Loading weights:  80%|████████  | 633/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.intermediate.dense.bias]         
Loading weights:  80%|████████  | 633/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.intermediate.dense.bias]
Loading weights:  80%|████████  | 634/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.intermediate.dense.weight]
Loading weights:  80%|████████  | 634/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.intermediate.dense.weight]
Loading weights:  81%|████████  | 635/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.LayerNorm.bias]    
Loading weights:  81%|████████  | 635/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.LayerNorm.bias]
Loading weights:  81%|████████  | 636/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  81%|████████  | 636/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.LayerNorm.weight]
Loading weights:  81%|████████  | 637/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.dense.bias]      
Loading weights:  81%|████████  | 637/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.dense.bias]
Loading weights:  81%|████████  | 638/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.dense.weight]
Loading weights:  81%|████████  | 638/788 [00:00<00:00, 1449.57it/s, Materializing param=text_encoder.encoder.layer.11.output.dense.weight]
Loading weights:  81%|████████  | 639/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.class_embedding]          
Loading weights:  81%|████████  | 639/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.class_embedding]
Loading weights:  81%|████████  | 640/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.patch_embedding.bias]
Loading weights:  81%|████████  | 640/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.patch_embedding.bias]
Loading weights:  81%|████████▏ | 641/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.patch_embedding.weight]
Loading weights:  81%|████████▏ | 641/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.patch_embedding.weight]
Loading weights:  81%|████████▏ | 642/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.position_embedding]    
Loading weights:  81%|████████▏ | 642/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.embeddings.position_embedding]
Loading weights:  82%|████████▏ | 643/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.bias]
Loading weights:  82%|████████▏ | 643/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.bias]
Loading weights:  82%|████████▏ | 644/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.weight]
Loading weights:  82%|████████▏ | 644/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm1.weight]
Loading weights:  82%|████████▏ | 645/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.bias]  
Loading weights:  82%|████████▏ | 645/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.bias]
Loading weights:  82%|████████▏ | 646/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.weight]
Loading weights:  82%|████████▏ | 646/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.layer_norm2.weight]
Loading weights:  82%|████████▏ | 647/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.bias]      
Loading weights:  82%|████████▏ | 647/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.bias]
Loading weights:  82%|████████▏ | 648/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.weight]
Loading weights:  82%|████████▏ | 648/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc1.weight]
Loading weights:  82%|████████▏ | 649/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.bias]  
Loading weights:  82%|████████▏ | 649/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.bias]
Loading weights:  82%|████████▏ | 650/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.weight]
Loading weights:  82%|████████▏ | 650/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.mlp.fc2.weight]
Loading weights:  83%|████████▎ | 651/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.bias]
Loading weights:  83%|████████▎ | 651/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.bias]
Loading weights:  83%|████████▎ | 652/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.weight]
Loading weights:  83%|████████▎ | 652/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.projection.weight]
Loading weights:  83%|████████▎ | 653/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.bias]         
Loading weights:  83%|████████▎ | 653/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.bias]
Loading weights:  83%|████████▎ | 654/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.weight]
Loading weights:  83%|████████▎ | 654/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.0.self_attn.qkv.weight]
Loading weights:  83%|████████▎ | 655/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.bias]    
Loading weights:  83%|████████▎ | 655/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.bias]
Loading weights:  83%|████████▎ | 656/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.weight]
Loading weights:  83%|████████▎ | 656/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm1.weight]
Loading weights:  83%|████████▎ | 657/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.bias]  
Loading weights:  83%|████████▎ | 657/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.bias]
Loading weights:  84%|████████▎ | 658/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.weight]
Loading weights:  84%|████████▎ | 658/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.layer_norm2.weight]
Loading weights:  84%|████████▎ | 659/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.bias]      
Loading weights:  84%|████████▎ | 659/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.bias]
Loading weights:  84%|████████▍ | 660/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.weight]
Loading weights:  84%|████████▍ | 660/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc1.weight]
Loading weights:  84%|████████▍ | 661/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.bias]  
Loading weights:  84%|████████▍ | 661/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.bias]
Loading weights:  84%|████████▍ | 662/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.weight]
Loading weights:  84%|████████▍ | 662/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.mlp.fc2.weight]
Loading weights:  84%|████████▍ | 663/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.bias]
Loading weights:  84%|████████▍ | 663/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.bias]
Loading weights:  84%|████████▍ | 664/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.weight]
Loading weights:  84%|████████▍ | 664/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.projection.weight]
Loading weights:  84%|████████▍ | 665/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.bias]         
Loading weights:  84%|████████▍ | 665/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.bias]
Loading weights:  85%|████████▍ | 666/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.weight]
Loading weights:  85%|████████▍ | 666/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.1.self_attn.qkv.weight]
Loading weights:  85%|████████▍ | 667/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.bias]    
Loading weights:  85%|████████▍ | 667/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.bias]
Loading weights:  85%|████████▍ | 668/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.weight]
Loading weights:  85%|████████▍ | 668/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm1.weight]
Loading weights:  85%|████████▍ | 669/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.bias]  
Loading weights:  85%|████████▍ | 669/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.bias]
Loading weights:  85%|████████▌ | 670/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.weight]
Loading weights:  85%|████████▌ | 670/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.layer_norm2.weight]
Loading weights:  85%|████████▌ | 671/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.bias]      
Loading weights:  85%|████████▌ | 671/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.bias]
Loading weights:  85%|████████▌ | 672/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.weight]
Loading weights:  85%|████████▌ | 672/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc1.weight]
Loading weights:  85%|████████▌ | 673/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.bias]  
Loading weights:  85%|████████▌ | 673/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.bias]
Loading weights:  86%|████████▌ | 674/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.weight]
Loading weights:  86%|████████▌ | 674/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.mlp.fc2.weight]
Loading weights:  86%|████████▌ | 675/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.bias]
Loading weights:  86%|████████▌ | 675/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.bias]
Loading weights:  86%|████████▌ | 676/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.weight]
Loading weights:  86%|████████▌ | 676/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.projection.weight]
Loading weights:  86%|████████▌ | 677/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.bias]         
Loading weights:  86%|████████▌ | 677/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.bias]
Loading weights:  86%|████████▌ | 678/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.weight]
Loading weights:  86%|████████▌ | 678/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.2.self_attn.qkv.weight]
Loading weights:  86%|████████▌ | 679/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.bias]    
Loading weights:  86%|████████▌ | 679/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.bias]
Loading weights:  86%|████████▋ | 680/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.weight]
Loading weights:  86%|████████▋ | 680/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm1.weight]
Loading weights:  86%|████████▋ | 681/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.bias]  
Loading weights:  86%|████████▋ | 681/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.bias]
Loading weights:  87%|████████▋ | 682/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.weight]
Loading weights:  87%|████████▋ | 682/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.layer_norm2.weight]
Loading weights:  87%|████████▋ | 683/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.bias]      
Loading weights:  87%|████████▋ | 683/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.bias]
Loading weights:  87%|████████▋ | 684/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.weight]
Loading weights:  87%|████████▋ | 684/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc1.weight]
Loading weights:  87%|████████▋ | 685/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.bias]  
Loading weights:  87%|████████▋ | 685/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.bias]
Loading weights:  87%|████████▋ | 686/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.weight]
Loading weights:  87%|████████▋ | 686/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.mlp.fc2.weight]
Loading weights:  87%|████████▋ | 687/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.bias]
Loading weights:  87%|████████▋ | 687/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.bias]
Loading weights:  87%|████████▋ | 688/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.weight]
Loading weights:  87%|████████▋ | 688/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.projection.weight]
Loading weights:  87%|████████▋ | 689/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.bias]         
Loading weights:  87%|████████▋ | 689/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.bias]
Loading weights:  88%|████████▊ | 690/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.weight]
Loading weights:  88%|████████▊ | 690/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.3.self_attn.qkv.weight]
Loading weights:  88%|████████▊ | 691/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.bias]    
Loading weights:  88%|████████▊ | 691/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.bias]
Loading weights:  88%|████████▊ | 692/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.weight]
Loading weights:  88%|████████▊ | 692/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm1.weight]
Loading weights:  88%|████████▊ | 693/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.bias]  
Loading weights:  88%|████████▊ | 693/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.bias]
Loading weights:  88%|████████▊ | 694/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.weight]
Loading weights:  88%|████████▊ | 694/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.layer_norm2.weight]
Loading weights:  88%|████████▊ | 695/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.bias]      
Loading weights:  88%|████████▊ | 695/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.bias]
Loading weights:  88%|████████▊ | 696/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.weight]
Loading weights:  88%|████████▊ | 696/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc1.weight]
Loading weights:  88%|████████▊ | 697/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.bias]  
Loading weights:  88%|████████▊ | 697/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.bias]
Loading weights:  89%|████████▊ | 698/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.weight]
Loading weights:  89%|████████▊ | 698/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.mlp.fc2.weight]
Loading weights:  89%|████████▊ | 699/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.bias]
Loading weights:  89%|████████▊ | 699/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.bias]
Loading weights:  89%|████████▉ | 700/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.weight]
Loading weights:  89%|████████▉ | 700/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.projection.weight]
Loading weights:  89%|████████▉ | 701/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.bias]         
Loading weights:  89%|████████▉ | 701/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.bias]
Loading weights:  89%|████████▉ | 702/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.weight]
Loading weights:  89%|████████▉ | 702/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.4.self_attn.qkv.weight]
Loading weights:  89%|████████▉ | 703/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.bias]    
Loading weights:  89%|████████▉ | 703/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.bias]
Loading weights:  89%|████████▉ | 704/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.weight]
Loading weights:  89%|████████▉ | 704/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm1.weight]
Loading weights:  89%|████████▉ | 705/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.bias]  
Loading weights:  89%|████████▉ | 705/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.bias]
Loading weights:  90%|████████▉ | 706/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.weight]
Loading weights:  90%|████████▉ | 706/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.layer_norm2.weight]
Loading weights:  90%|████████▉ | 707/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.bias]      
Loading weights:  90%|████████▉ | 707/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.bias]
Loading weights:  90%|████████▉ | 708/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.weight]
Loading weights:  90%|████████▉ | 708/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc1.weight]
Loading weights:  90%|████████▉ | 709/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.bias]  
Loading weights:  90%|████████▉ | 709/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.bias]
Loading weights:  90%|█████████ | 710/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.weight]
Loading weights:  90%|█████████ | 710/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.mlp.fc2.weight]
Loading weights:  90%|█████████ | 711/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.bias]
Loading weights:  90%|██��██████ | 711/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.bias]
Loading weights:  90%|█████████ | 712/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.weight]
Loading weights:  90%|█████████ | 712/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.projection.weight]
Loading weights:  90%|█████████ | 713/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.bias]         
Loading weights:  90%|█████████ | 713/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.bias]
Loading weights:  91%|█████████ | 714/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.weight]
Loading weights:  91%|█████████ | 714/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.5.self_attn.qkv.weight]
Loading weights:  91%|█████████ | 715/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.bias]    
Loading weights:  91%|█████████ | 715/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.bias]
Loading weights:  91%|█████████ | 716/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.weight]
Loading weights:  91%|█████████ | 716/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm1.weight]
Loading weights:  91%|█████████ | 717/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.bias]  
Loading weights:  91%|█████████ | 717/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.bias]
Loading weights:  91%|█████████ | 718/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.weight]
Loading weights:  91%|█████████ | 718/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.layer_norm2.weight]
Loading weights:  91%|█████████ | 719/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.bias]      
Loading weights:  91%|█████████ | 719/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.bias]
Loading weights:  91%|█████████▏| 720/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.weight]
Loading weights:  91%|█████████▏| 720/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc1.weight]
Loading weights:  91%|█████████▏| 721/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.bias]  
Loading weights:  91%|█████████▏| 721/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.bias]
Loading weights:  92%|█████████▏| 722/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.weight]
Loading weights:  92%|█████████▏| 722/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.mlp.fc2.weight]
Loading weights:  92%|█████████▏| 723/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.bias]
Loading weights:  92%|█████████▏| 723/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.bias]
Loading weights:  92%|█████████▏| 724/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.weight]
Loading weights:  92%|█████████▏| 724/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.projection.weight]
Loading weights:  92%|█████████▏| 725/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.bias]         
Loading weights:  92%|█████████▏| 725/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.bias]
Loading weights:  92%|█████████▏| 726/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.weight]
Loading weights:  92%|█████████▏| 726/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.6.self_attn.qkv.weight]
Loading weights:  92%|█████████▏| 727/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.bias]    
Loading weights:  92%|███████��█▏| 727/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.bias]
Loading weights:  92%|█████████▏| 728/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.weight]
Loading weights:  92%|█████████▏| 728/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm1.weight]
Loading weights:  93%|█████████▎| 729/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.bias]  
Loading weights:  93%|█████████▎| 729/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.bias]
Loading weights:  93%|█████████▎| 730/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.weight]
Loading weights:  93%|█████████▎| 730/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.layer_norm2.weight]
Loading weights:  93%|█████████▎| 731/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.bias]      
Loading weights:  93%|█████████▎| 731/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.bias]
Loading weights:  93%|█████████▎| 732/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.weight]
Loading weights:  93%|█████████▎| 732/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc1.weight]
Loading weights:  93%|█████████▎| 733/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.bias]  
Loading weights:  93%|█████████▎| 733/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.bias]
Loading weights:  93%|█████████▎| 734/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.weight]
Loading weights:  93%|█████████▎| 734/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.mlp.fc2.weight]
Loading weights:  93%|█████████▎| 735/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.bias]
Loading weights:  93%|█████████▎| 735/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.bias]
Loading weights:  93%|█████████▎| 736/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.weight]
Loading weights:  93%|█████████▎| 736/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.projection.weight]
Loading weights:  94%|█████████▎| 737/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.bias]         
Loading weights:  94%|█████████▎| 737/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.bias]
Loading weights:  94%|█████████▎| 738/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.weight]
Loading weights:  94%|█████████▎| 738/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.7.self_attn.qkv.weight]
Loading weights:  94%|█████████▍| 739/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.bias]    
Loading weights:  94%|█████████▍| 739/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.bias]
Loading weights:  94%|█████████▍| 740/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.weight]
Loading weights:  94%|█████████▍| 740/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm1.weight]
Loading weights:  94%|█████████▍| 741/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.bias]  
Loading weights:  94%|█████████▍| 741/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.bias]
Loading weights:  94%|█████████▍| 742/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.weight]
Loading weights:  94%|█████████▍| 742/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.layer_norm2.weight]
Loading weights:  94%|█████████▍| 743/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.bias]      
Loading weights:  94%|█████████▍| 743/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.bias]
Loading weights:  94%|█████████▍| 744/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.weight]
Loading weights:  94%|█████████▍| 744/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc1.weight]
Loading weights:  95%|█████████▍| 745/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.bias]  
Loading weights:  95%|█████████▍| 745/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.bias]
Loading weights:  95%|█████████▍| 746/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.weight]
Loading weights:  95%|█████████▍| 746/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.mlp.fc2.weight]
Loading weights:  95%|█████████▍| 747/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.bias]
Loading weights:  95%|█████████▍| 747/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.bias]
Loading weights:  95%|█████████▍| 748/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.weight]
Loading weights:  95%|█████████▍| 748/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.projection.weight]
Loading weights:  95%|█████████▌| 749/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.bias]         
Loading weights:  95%|█████████▌| 749/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.bias]
Loading weights:  95%|█████████▌| 750/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.weight]
Loading weights:  95%|█████████▌| 750/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.8.self_attn.qkv.weight]
Loading weights:  95%|█████████▌| 751/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.bias]    
Loading weights:  95%|█████████▌| 751/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.bias]
Loading weights:  95%|█████████▌| 752/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.weight]
Loading weights:  95%|█████████▌| 752/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm1.weight]
Loading weights:  96%|█████████▌| 753/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.bias]  
Loading weights:  96%|█████████▌| 753/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.bias]
Loading weights:  96%|█████████▌| 754/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.weight]
Loading weights:  96%|█████████▌| 754/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.layer_norm2.weight]
Loading weights:  96%|█████████▌| 755/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.bias]      
Loading weights:  96%|█████████▌| 755/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.bias]
Loading weights:  96%|█████████▌| 756/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.weight]
Loading weights:  96%|█████████▌| 756/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc1.weight]
Loading weights:  96%|█████████▌| 757/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.bias]  
Loading weights:  96%|█████████▌| 757/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.bias]
Loading weights:  96%|█████████▌| 758/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.weight]
Loading weights:  96%|█████████▌| 758/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.mlp.fc2.weight]
Loading weights:  96%|█████████▋| 759/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.bias]
Loading weights:  96%|█████████▋| 759/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.bias]
Loading weights:  96%|█████████▋| 760/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.weight]
Loading weights:  96%|█████████▋| 760/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.projection.weight]
Loading weights:  97%|█████████▋| 761/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.bias]         
Loading weights:  97%|█████████▋| 761/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.bias]
Loading weights:  97%|█████████▋| 762/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.weight]
Loading weights:  97%|█████████▋| 762/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.9.self_attn.qkv.weight]
Loading weights:  97%|█████████▋| 763/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.bias]   
Loading weights:  97%|█████████▋| 763/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.bias]
Loading weights:  97%|█████████▋| 764/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.weight]
Loading weights:  97%|█████████▋| 764/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm1.weight]
Loading weights:  97%|█████████▋| 765/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.bias]  
Loading weights:  97%|█████████▋| 765/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.bias]
Loading weights:  97%|█████████▋| 766/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.weight]
Loading weights:  97%|█████████▋| 766/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.layer_norm2.weight]
Loading weights:  97%|█████████▋| 767/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.bias]      
Loading weights:  97%|█████████▋| 767/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.bias]
Loading weights:  97%|█████████▋| 768/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.weight]
Loading weights:  97%|█████████▋| 768/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc1.weight]
Loading weights:  98%|█████████▊| 769/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.bias]  
Loading weights:  98%|█████████▊| 769/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.bias]
Loading weights:  98%|█████████▊| 770/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.weight]
Loading weights:  98%|█████████▊| 770/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.mlp.fc2.weight]
Loading weights:  98%|█████████▊| 771/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.bias]
Loading weights:  98%|█████████▊| 771/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.bias]
Loading weights:  98%|█████████▊| 772/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.weight]
Loading weights:  98%|█████████▊| 772/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.projection.weight]
Loading weights:  98%|█████████▊| 773/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.bias]         
Loading weights:  98%|█████████▊| 773/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.bias]
Loading weights:  98%|█████████▊| 774/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.weight]
Loading weights:  98%|█████████▊| 774/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.10.self_attn.qkv.weight]
Loading weights:  98%|█████████▊| 775/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.bias]    
Loading weights:  98%|█████████▊| 775/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.bias]
Loading weights:  98%|█████████▊| 776/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.weight]
Loading weights:  98%|█████████▊| 776/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm1.weight]
Loading weights:  99%|█████████▊| 777/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.bias]  
Loading weights:  99%|█████████▊| 777/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.bias]
Loading weights:  99%|█████████▊| 778/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.weight]
Loading weights:  99%|█████████▊| 778/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.layer_norm2.weight]
Loading weights:  99%|█████████▉| 779/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.bias]      
Loading weights:  99%|█████████▉| 779/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.bias]
Loading weights:  99%|█████████▉| 780/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.weight]
Loading weights:  99%|█████████▉| 780/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc1.weight]
Loading weights:  99%|█████████▉| 781/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.bias]  
Loading weights:  99%|█████████▉| 781/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.bias]
Loading weights:  99%|█████████▉| 782/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.weight]
Loading weights:  99%|█████████▉| 782/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.mlp.fc2.weight]
Loading weights:  99%|█████████▉| 783/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.bias]
Loading weights:  99%|█████████▉| 783/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.bias]
Loading weights:  99%|█████████▉| 784/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.weight]
Loading weights:  99%|█████████▉| 784/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.projection.weight]
Loading weights: 100%|█████████▉| 785/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.bias]         
Loading weights: 100%|█████████▉| 785/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.bias]
Loading weights: 100%|█████████▉| 786/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.weight]
Loading weights: 100%|█████████▉| 786/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.encoder.layers.11.self_attn.qkv.weight]
Loading weights: 100%|█████████▉| 787/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.post_layernorm.bias]                   
Loading weights: 100%|█████████▉| 787/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.post_layernorm.bias]
Loading weights: 100%|██████████| 788/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.post_layernorm.weight]
Loading weights: 100%|██████████| 788/788 [00:00<00:00, 1449.57it/s, Materializing param=vision_model.post_layernorm.weight]
Loading weights: 100%|██████████| 788/788 [00:00<00:00, 5527.35it/s, Materializing param=vision_model.post_layernorm.weight]
+The tied weights mapping and config for this model specifies to tie text_decoder.bert.embeddings.word_embeddings.weight to text_decoder.cls.predictions.decoder.weight, but both are present in the checkpoints, so we will NOT tie them. You should update the config with `tie_word_embeddings=False` to silence this warning
+BlipForQuestionAnswering LOAD REPORT from: Salesforce/blip-vqa-base
+Key                                       | Status     |  | 
+------------------------------------------+------------+--+-
+text_encoder.embeddings.position_ids      | UNEXPECTED |  | 
+text_decoder.bert.embeddings.position_ids | UNEXPECTED |  | 
+
+Notes:
+- UNEXPECTED	:can be ignored when loading from different task/architecture; not ok if you expect identical arch.
+  ✓ BLIP (captioning + VQA)
+📚 Loading COCO data for reasoning training...
+✅ Loaded 22840 reasoning samples
+   - Captions: 10000
+   - Existence: 9912
+   - Counting: 2928
+
+📊 Training Config:
+   Samples: 22840
+   Batch size: 8
+   Epochs: 8
+   Total steps: 22840
+   LR: 3e-5 -> 1e-6 (cosine)
+
+🔥 Starting training...
+
Epoch 1/8:   0%|          | 0/2855 [00:00<?, ?it/s]We strongly recommend passing in an `attention_mask` since your input_ids may be padded. See https://huggingface.co/docs/transformers/troubleshooting#incorrect-output-when-padding-tokens-arent-masked.
+
Epoch 1/8:   0%|          | 0/2855 [00:06<?, ?it/s, loss=10.3244, lr=3.00e-05]
Epoch 1/8:   0%|          | 1/2855 [00:06<4:57:06,  6.25s/it, loss=10.3244, lr=3.00e-05]
Epoch 1/8:   0%|          | 1/2855 [00:11<4:57:06,  6.25s/it, loss=9.8016, lr=3.00e-05] 
Epoch 1/8:   0%|          | 2/2855 [00:11<4:40:40,  5.90s/it, loss=9.8016, lr=3.00e-05]
Epoch 1/8:   0%|          | 2/2855 [00:17<4:40:40,  5.90s/it, loss=9.9271, lr=3.00e-05]
Epoch 1/8:   0%|          | 3/2855 [00:17<4:40:52,  5.91s/it, loss=9.9271, lr=3.00e-05]
Epoch 1/8:   0%|          | 3/2855 [00:21<4:40:52,  5.91s/it, loss=9.5596, lr=3.00e-05]
Epoch 1/8:   0%|          | 4/2855 [00:21<3:54:48,  4.94s/it, loss=9.5596, lr=3.00e-05]
Epoch 1/8:   0%|          | 4/2855 [00:23<3:54:48,  4.94s/it, loss=9.6195, lr=3.00e-05]
Epoch 1/8:   0%|          | 5/2855 [00:23<3:03:00,  3.85s/it, loss=9.6195, lr=3.00e-05]
Epoch 1/8:   0%|          | 5/2855 [00:26<3:03:00,  3.85s/it, loss=9.3526, lr=3.00e-05]
Epoch 1/8:   0%|          | 6/2855 [00:26<2:47:32,  3.53s/it, loss=9.3526, lr=3.00e-05]
Epoch 1/8:   0%|          | 6/2855 [00:34<2:47:32,  3.53s/it, loss=9.7432, lr=3.00e-05]
Epoch 1/8:   0%|          | 7/2855 [00:34<3:59:18,  5.04s/it, loss=9.7432, lr=3.00e-05]
Epoch 1/8:   0%|          | 7/2855 [00:41<3:59:18,  5.04s/it, loss=9.8075, lr=3.00e-05]
Epoch 1/8:   0%|          | 8/2855 [00:41<4:31:12,  5.72s/it, loss=9.8075, lr=3.00e-05]
Epoch 1/8:   0%|          | 8/2855 [00:45<4:31:12,  5.72s/it, loss=8.8199, lr=3.00e-05]
Epoch 1/8:   0%|          | 9/2855 [00:45<4:06:23,  5.19s/it, loss=8.8199, lr=3.00e-05]
Epoch 1/8:   0%|          | 9/2855 [00:49<4:06:23,  5.19s/it, loss=9.3081, lr=3.00e-05]
Epoch 1/8:   0%|          | 10/2855 [00:49<3:44:48,  4.74s/it, loss=9.3081, lr=3.00e-05]
Epoch 1/8:   0%|          | 10/2855 [00:52<3:44:48,  4.74s/it, loss=9.0872, lr=3.00e-05]
Epoch 1/8:   0%|          | 11/2855 [00:52<3:30:18,  4.44s/it, loss=9.0872, lr=3.00e-05]
Epoch 1/8:   0%|          | 11/2855 [00:56<3:30:18,  4.44s/it, loss=9.0806, lr=3.00e-05]
Epoch 1/8:   0%|          | 12/2855 [00:56<3:16:40,  4.15s/it, loss=9.0806, lr=3.00e-05]
Epoch 1/8:   0%|          | 12/2855 [01:00<3:16:40,  4.15s/it, loss=8.5038, lr=3.00e-05]
Epoch 1/8:   0%|          | 13/2855 [01:00<3:11:03,  4.03s/it, loss=8.5038, lr=3.00e-05]
Epoch 1/8:   0%|          | 13/2855 [01:03<3:11:03,  4.03s/it, loss=8.7113, lr=3.00e-05]
Epoch 1/8:   0%|          | 14/2855 [01:03<3:07:05,  3.95s/it, loss=8.7113, lr=3.00e-05]
Epoch 1/8:   0%|          | 14/2855 [01:07<3:07:05,  3.95s/it, loss=8.6661, lr=3.00e-05]
Epoch 1/8:   1%|          | 15/2855 [01:07<2:55:49,  3.71s/it, loss=8.6661, lr=3.00e-05]
Epoch 1/8:   1%|          | 15/2855 [01:10<2:55:49,  3.71s/it, loss=8.4461, lr=3.00e-05]
Epoch 1/8:   1%|          | 16/2855 [01:10<2:54:00,  3.68s/it, loss=8.4461, lr=3.00e-05]
Epoch 1/8:   1%|          | 16/2855 [01:13<2:54:00,  3.68s/it, loss=8.3518, lr=3.00e-05]
Epoch 1/8:   1%|          | 17/2855 [01:13<2:44:53,  3.49s/it, loss=8.3518, lr=3.00e-05]
Epoch 1/8:   1%|          | 17/2855 [01:16<2:44:53,  3.49s/it, loss=8.3166, lr=3.00e-05]
Epoch 1/8:   1%|          | 18/2855 [01:16<2:39:02,  3.36s/it, loss=8.3166, lr=3.00e-05]
Epoch 1/8:   1%|          | 18/2855 [01:20<2:39:02,  3.36s/it, loss=8.3217, lr=3.00e-05]
Epoch 1/8:   1%|          | 19/2855 [01:20<2:42:45,  3.44s/it, loss=8.3217, lr=3.00e-05]
Epoch 1/8:   1%|          | 19/2855 [01:23<2:42:45,  3.44s/it, loss=8.0889, lr=3.00e-05]
Epoch 1/8:   1%|          | 20/2855 [01:23<2:39:53,  3.38s/it, loss=8.0889, lr=3.00e-05]
Epoch 1/8:   1%|          | 20/2855 [01:27<2:39:53,  3.38s/it, loss=8.0973, lr=3.00e-05]
Epoch 1/8:   1%|          | 21/2855 [01:27<2:39:26,  3.38s/it, loss=8.0973, lr=3.00e-05]
Epoch 1/8:   1%|          | 21/2855 [01:30<2:39:26,  3.38s/it, loss=8.2244, lr=3.00e-05]
Epoch 1/8:   1%|          | 22/2855 [01:30<2:37:15,  3.33s/it, loss=8.2244, lr=3.00e-05]
Epoch 1/8:   1%|          | 22/2855 [01:33<2:37:15,  3.33s/it, loss=8.0312, lr=3.00e-05]
Epoch 1/8:   1%|          | 23/2855 [01:33<2:35:48,  3.30s/it, loss=8.0312, lr=3.00e-05]
Epoch 1/8:   1%|          | 23/2855 [01:39<2:35:48,  3.30s/it, loss=7.8560, lr=3.00e-05]
Epoch 1/8:   1%|          | 24/2855 [01:39<3:07:25,  3.97s/it, loss=7.8560, lr=3.00e-05]
Epoch 1/8:   1%|          | 24/2855 [01:48<3:07:25,  3.97s/it, loss=7.9000, lr=3.00e-05]
Epoch 1/8:   1%|          | 25/2855 [01:48<4:17:59,  5.47s/it, loss=7.9000, lr=3.00e-05]
Epoch 1/8:   1%|          | 25/2855 [03:26<4:17:59,  5.47s/it, loss=7.5780, lr=3.00e-05]
Epoch 1/8:   1%|          | 26/2855 [03:26<26:12:27, 33.35s/it, loss=7.5780, lr=3.00e-05]
Epoch 1/8:   1%|          | 26/2855 [03:28<26:12:27, 33.35s/it, loss=7.6876, lr=3.00e-05]
Epoch 1/8:   1%|          | 27/2855 [03:28<18:53:46, 24.05s/it, loss=7.6876, lr=3.00e-05]
Epoch 1/8:   1%|          | 27/2855 [03:31<18:53:46, 24.05s/it, loss=7.4089, lr=3.00e-05]
Epoch 1/8:   1%|          | 28/2855 [03:31<13:52:30, 17.67s/it, loss=7.4089, lr=3.00e-05]
Epoch 1/8:   1%|          | 28/2855 [03:33<13:52:30, 17.67s/it, loss=7.3499, lr=3.00e-05]
Epoch 1/8:   1%|          | 29/2855 [03:33<10:16:08, 13.08s/it, loss=7.3499, lr=3.00e-05]
Epoch 1/8:   1%|          | 29/2855 [03:36<10:16:08, 13.08s/it, loss=7.3827, lr=3.00e-05]
Epoch 1/8:   1%|          | 30/2855 [03:36<7:43:20,  9.84s/it, loss=7.3827, lr=3.00e-05] 
Epoch 1/8:   1%|          | 30/2855 [03:38<7:43:20,  9.84s/it, loss=7.3478, lr=3.00e-05]
Epoch 1/8:   1%|          | 31/2855 [03:38<6:01:08,  7.67s/it, loss=7.3478, lr=3.00e-05]
Epoch 1/8:   1%|          | 31/2855 [03:40<6:01:08,  7.67s/it, loss=7.4301, lr=3.00e-05]
Epoch 1/8:   1%|          | 32/2855 [03:40<4:43:00,  6.02s/it, loss=7.4301, lr=3.00e-05]
Epoch 1/8:   1%|          | 32/2855 [03:43<4:43:00,  6.02s/it, loss=7.0278, lr=3.00e-05]
Epoch 1/8:   1%|          | 33/2855 [03:43<3:57:49,  5.06s/it, loss=7.0278, lr=3.00e-05]
Epoch 1/8:   1%|          | 33/2855 [03:45<3:57:49,  5.06s/it, loss=6.9140, lr=3.00e-05]
Epoch 1/8:   1%|          | 34/2855 [03:45<3:16:46,  4.19s/it, loss=6.9140, lr=3.00e-05]
Epoch 1/8:   1%|          | 34/2855 [03:48<3:16:46,  4.19s/it, loss=6.9058, lr=3.00e-05]
Epoch 1/8:   1%|          | 35/2855 [03:48<2:53:47,  3.70s/it, loss=6.9058, lr=3.00e-05]
Epoch 1/8:   1%|          | 35/2855 [03:50<2:53:47,  3.70s/it, loss=6.9507, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 36/2855 [03:50<2:30:10,  3.20s/it, loss=6.9507, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 36/2855 [03:52<2:30:10,  3.20s/it, loss=6.7045, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 37/2855 [03:52<2:18:06,  2.94s/it, loss=6.7045, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 37/2855 [03:54<2:18:06,  2.94s/it, loss=6.6722, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 38/2855 [03:54<2:04:14,  2.65s/it, loss=6.6722, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 38/2855 [03:56<2:04:14,  2.65s/it, loss=6.7147, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 39/2855 [03:56<1:56:47,  2.49s/it, loss=6.7147, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 39/2855 [03:59<1:56:47,  2.49s/it, loss=6.4643, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 40/2855 [03:59<1:56:38,  2.49s/it, loss=6.4643, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 40/2855 [04:01<1:56:38,  2.49s/it, loss=6.5806, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 41/2855 [04:01<1:51:07,  2.37s/it, loss=6.5806, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 41/2855 [04:03<1:51:07,  2.37s/it, loss=6.5055, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 42/2855 [04:03<1:52:06,  2.39s/it, loss=6.5055, lr=3.00e-05]
Epoch 1/8:   1%|▏         | 42/2855 [04:06<1:52:06,  2.39s/it, loss=6.4100, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 43/2855 [04:06<1:48:41,  2.32s/it, loss=6.4100, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 43/2855 [04:08<1:48:41,  2.32s/it, loss=6.3852, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 44/2855 [04:08<1:54:41,  2.45s/it, loss=6.3852, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 44/2855 [04:11<1:54:41,  2.45s/it, loss=6.2786, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 45/2855 [04:11<1:51:38,  2.38s/it, loss=6.2786, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 45/2855 [04:13<1:51:38,  2.38s/it, loss=6.1235, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 46/2855 [04:13<1:55:08,  2.46s/it, loss=6.1235, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 46/2855 [04:15<1:55:08,  2.46s/it, loss=6.1129, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 47/2855 [04:15<1:50:40,  2.36s/it, loss=6.1129, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 47/2855 [04:18<1:50:40,  2.36s/it, loss=6.0954, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 48/2855 [04:18<1:51:48,  2.39s/it, loss=6.0954, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 48/2855 [04:20<1:51:48,  2.39s/it, loss=6.0222, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 49/2855 [04:20<1:48:34,  2.32s/it, loss=6.0222, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 49/2855 [04:22<1:48:34,  2.32s/it, loss=5.8123, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 50/2855 [04:22<1:47:45,  2.31s/it, loss=5.8123, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 50/2855 [04:25<1:47:45,  2.31s/it, loss=5.6860, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 51/2855 [04:25<1:48:21,  2.32s/it, loss=5.6860, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 51/2855 [04:27<1:48:21,  2.32s/it, loss=5.7100, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 52/2855 [04:27<1:45:09,  2.25s/it, loss=5.7100, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 52/2855 [04:29<1:45:09,  2.25s/it, loss=5.7634, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 53/2855 [04:29<1:48:56,  2.33s/it, loss=5.7634, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 53/2855 [04:31<1:48:56,  2.33s/it, loss=5.5021, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 54/2855 [04:31<1:45:45,  2.27s/it, loss=5.5021, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 54/2855 [04:34<1:45:45,  2.27s/it, loss=5.4493, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 55/2855 [04:34<1:47:25,  2.30s/it, loss=5.4493, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 55/2855 [04:36<1:47:25,  2.30s/it, loss=5.2967, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 56/2855 [04:36<1:43:05,  2.21s/it, loss=5.2967, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 56/2855 [04:38<1:43:05,  2.21s/it, loss=5.4748, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 57/2855 [04:38<1:47:10,  2.30s/it, loss=5.4748, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 57/2855 [04:40<1:47:10,  2.30s/it, loss=5.0946, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 58/2855 [04:40<1:45:34,  2.26s/it, loss=5.0946, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 58/2855 [04:43<1:45:34,  2.26s/it, loss=5.1783, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 59/2855 [04:43<1:55:09,  2.47s/it, loss=5.1783, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 59/2855 [04:45<1:55:09,  2.47s/it, loss=5.1007, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 60/2855 [04:45<1:49:17,  2.35s/it, loss=5.1007, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 60/2855 [04:52<1:49:17,  2.35s/it, loss=5.1448, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 61/2855 [04:52<2:53:45,  3.73s/it, loss=5.1448, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 61/2855 [04:56<2:53:45,  3.73s/it, loss=5.0831, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 62/2855 [04:56<2:50:12,  3.66s/it, loss=5.0831, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 62/2855 [05:04<2:50:12,  3.66s/it, loss=4.9177, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 63/2855 [05:04<3:57:52,  5.11s/it, loss=4.9177, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 63/2855 [05:08<3:57:52,  5.11s/it, loss=4.9051, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 64/2855 [05:08<3:30:41,  4.53s/it, loss=4.9051, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 64/2855 [05:10<3:30:41,  4.53s/it, loss=4.8667, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 65/2855 [05:10<3:02:24,  3.92s/it, loss=4.8667, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 65/2855 [05:13<3:02:24,  3.92s/it, loss=4.8608, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 66/2855 [05:13<2:42:01,  3.49s/it, loss=4.8608, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 66/2855 [05:15<2:42:01,  3.49s/it, loss=4.9436, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 67/2855 [05:15<2:26:36,  3.16s/it, loss=4.9436, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 67/2855 [05:17<2:26:36,  3.16s/it, loss=4.5184, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 68/2855 [05:17<2:16:34,  2.94s/it, loss=4.5184, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 68/2855 [05:20<2:16:34,  2.94s/it, loss=4.5162, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 69/2855 [05:20<2:11:11,  2.83s/it, loss=4.5162, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 69/2855 [05:22<2:11:11,  2.83s/it, loss=4.4306, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 70/2855 [05:22<2:03:04,  2.65s/it, loss=4.4306, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 70/2855 [05:25<2:03:04,  2.65s/it, loss=4.4997, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 71/2855 [05:25<2:00:10,  2.59s/it, loss=4.4997, lr=3.00e-05]
Epoch 1/8:   2%|▏         | 71/2855 [05:27<2:00:10,  2.59s/it, loss=4.2953, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 72/2855 [05:27<1:54:14,  2.46s/it, loss=4.2953, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 72/2855 [05:30<1:54:14,  2.46s/it, loss=4.2175, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 73/2855 [05:30<1:58:13,  2.55s/it, loss=4.2175, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 73/2855 [05:32<1:58:13,  2.55s/it, loss=4.1765, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 74/2855 [05:32<1:51:17,  2.40s/it, loss=4.1765, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 74/2855 [05:34<1:51:17,  2.40s/it, loss=4.1809, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 75/2855 [05:34<1:53:14,  2.44s/it, loss=4.1809, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 75/2855 [05:37<1:53:14,  2.44s/it, loss=4.0705, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 76/2855 [05:37<2:04:18,  2.68s/it, loss=4.0705, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 76/2855 [05:45<2:04:18,  2.68s/it, loss=3.9312, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 77/2855 [05:45<3:19:33,  4.31s/it, loss=3.9312, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 77/2855 [05:48<3:19:33,  4.31s/it, loss=3.9505, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 78/2855 [05:48<2:52:46,  3.73s/it, loss=3.9505, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 78/2855 [05:51<2:52:46,  3.73s/it, loss=3.8224, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 79/2855 [05:51<2:37:46,  3.41s/it, loss=3.8224, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 79/2855 [05:53<2:37:46,  3.41s/it, loss=3.8140, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 80/2855 [05:53<2:19:57,  3.03s/it, loss=3.8140, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 80/2855 [05:55<2:19:57,  3.03s/it, loss=3.8674, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 81/2855 [05:55<2:14:04,  2.90s/it, loss=3.8674, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 81/2855 [05:57<2:14:04,  2.90s/it, loss=3.6431, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 82/2855 [05:57<2:03:24,  2.67s/it, loss=3.6431, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 82/2855 [06:00<2:03:24,  2.67s/it, loss=3.6327, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 83/2855 [06:00<2:07:49,  2.77s/it, loss=3.6327, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 83/2855 [06:06<2:07:49,  2.77s/it, loss=3.6188, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 84/2855 [06:06<2:43:23,  3.54s/it, loss=3.6188, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 84/2855 [06:08<2:43:23,  3.54s/it, loss=3.5734, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 85/2855 [06:08<2:29:42,  3.24s/it, loss=3.5734, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 85/2855 [06:11<2:29:42,  3.24s/it, loss=3.3175, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 86/2855 [06:11<2:18:31,  3.00s/it, loss=3.3175, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 86/2855 [06:17<2:18:31,  3.00s/it, loss=3.2188, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 87/2855 [06:17<3:02:05,  3.95s/it, loss=3.2188, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 87/2855 [06:20<3:02:05,  3.95s/it, loss=3.2638, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 88/2855 [06:20<2:46:16,  3.61s/it, loss=3.2638, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 88/2855 [06:22<2:46:16,  3.61s/it, loss=3.1416, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 89/2855 [06:22<2:24:41,  3.14s/it, loss=3.1416, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 89/2855 [06:24<2:24:41,  3.14s/it, loss=3.1719, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 90/2855 [06:24<2:10:48,  2.84s/it, loss=3.1719, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 90/2855 [06:26<2:10:48,  2.84s/it, loss=3.1179, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 91/2855 [06:26<2:01:30,  2.64s/it, loss=3.1179, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 91/2855 [06:28<2:01:30,  2.64s/it, loss=3.0304, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 92/2855 [06:28<1:56:34,  2.53s/it, loss=3.0304, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 92/2855 [06:31<1:56:34,  2.53s/it, loss=2.9526, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 93/2855 [06:31<1:52:11,  2.44s/it, loss=2.9526, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 93/2855 [06:33<1:52:11,  2.44s/it, loss=2.8199, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 94/2855 [06:33<1:48:08,  2.35s/it, loss=2.8199, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 94/2855 [06:36<1:48:08,  2.35s/it, loss=2.7764, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 95/2855 [06:36<1:59:28,  2.60s/it, loss=2.7764, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 95/2855 [06:38<1:59:28,  2.60s/it, loss=2.7504, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 96/2855 [06:38<1:53:26,  2.47s/it, loss=2.7504, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 96/2855 [06:40<1:53:26,  2.47s/it, loss=2.5895, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 97/2855 [06:40<1:53:34,  2.47s/it, loss=2.5895, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 97/2855 [06:43<1:53:34,  2.47s/it, loss=2.5641, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 98/2855 [06:43<1:51:17,  2.42s/it, loss=2.5641, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 98/2855 [06:45<1:51:17,  2.42s/it, loss=2.5367, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 99/2855 [06:45<1:47:13,  2.33s/it, loss=2.5367, lr=3.00e-05]
Epoch 1/8:   3%|▎         | 99/2855 [06:47<1:47:13,  2.33s/it, loss=2.4022, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 100/2855 [06:47<1:46:32,  2.32s/it, loss=2.4022, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 100/2855 [06:49<1:46:32,  2.32s/it, loss=2.4261, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 101/2855 [06:49<1:44:42,  2.28s/it, loss=2.4261, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 101/2855 [06:51<1:44:42,  2.28s/it, loss=2.3488, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 102/2855 [06:51<1:41:59,  2.22s/it, loss=2.3488, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 102/2855 [06:53<1:41:59,  2.22s/it, loss=2.2651, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 103/2855 [06:53<1:38:33,  2.15s/it, loss=2.2651, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 103/2855 [06:56<1:38:33,  2.15s/it, loss=2.2727, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 104/2855 [06:56<1:44:13,  2.27s/it, loss=2.2727, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 104/2855 [06:58<1:44:13,  2.27s/it, loss=2.1174, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 105/2855 [06:58<1:42:12,  2.23s/it, loss=2.1174, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 105/2855 [07:00<1:42:12,  2.23s/it, loss=2.0973, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 106/2855 [07:00<1:42:05,  2.23s/it, loss=2.0973, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 106/2855 [07:03<1:42:05,  2.23s/it, loss=1.9931, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 107/2855 [07:03<1:43:06,  2.25s/it, loss=1.9931, lr=3.00e-05]
Epoch 1/8:   4%|▎         | 107/2855 [07:05<1:43:06,  2.25s/it, loss=2.1112, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 108/2855 [07:05<1:48:14,  2.36s/it, loss=2.1112, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 108/2855 [07:07<1:48:14,  2.36s/it, loss=1.9241, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 109/2855 [07:07<1:44:38,  2.29s/it, loss=1.9241, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 109/2855 [07:10<1:44:38,  2.29s/it, loss=1.9727, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 110/2855 [07:10<1:45:02,  2.30s/it, loss=1.9727, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 110/2855 [07:12<1:45:02,  2.30s/it, loss=1.9102, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 111/2855 [07:12<1:44:01,  2.27s/it, loss=1.9102, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 111/2855 [07:14<1:44:01,  2.27s/it, loss=1.7849, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 112/2855 [07:14<1:41:15,  2.21s/it, loss=1.7849, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 112/2855 [07:16<1:41:15,  2.21s/it, loss=1.6114, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 113/2855 [07:16<1:40:31,  2.20s/it, loss=1.6114, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 113/2855 [07:18<1:40:31,  2.20s/it, loss=1.6801, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 114/2855 [07:18<1:39:19,  2.17s/it, loss=1.6801, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 114/2855 [07:20<1:39:19,  2.17s/it, loss=1.6650, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 115/2855 [07:20<1:39:14,  2.17s/it, loss=1.6650, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 115/2855 [07:23<1:39:14,  2.17s/it, loss=1.5051, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 116/2855 [07:23<1:38:41,  2.16s/it, loss=1.5051, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 116/2855 [07:25<1:38:41,  2.16s/it, loss=1.5630, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 117/2855 [07:25<1:38:29,  2.16s/it, loss=1.5630, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 117/2855 [07:27<1:38:29,  2.16s/it, loss=1.5345, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 118/2855 [07:27<1:39:11,  2.17s/it, loss=1.5345, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 118/2855 [07:29<1:39:11,  2.17s/it, loss=1.4736, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 119/2855 [07:29<1:38:10,  2.15s/it, loss=1.4736, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 119/2855 [07:31<1:38:10,  2.15s/it, loss=1.2662, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 120/2855 [07:31<1:37:56,  2.15s/it, loss=1.2662, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 120/2855 [07:33<1:37:56,  2.15s/it, loss=1.1962, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 121/2855 [07:33<1:36:55,  2.13s/it, loss=1.1962, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 121/2855 [07:35<1:36:55,  2.13s/it, loss=1.3512, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 122/2855 [07:35<1:37:01,  2.13s/it, loss=1.3512, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 122/2855 [07:38<1:37:01,  2.13s/it, loss=1.3033, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 123/2855 [07:38<1:37:59,  2.15s/it, loss=1.3033, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 123/2855 [07:40<1:37:59,  2.15s/it, loss=1.2081, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 124/2855 [07:40<1:35:48,  2.10s/it, loss=1.2081, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 124/2855 [07:42<1:35:48,  2.10s/it, loss=1.2716, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 125/2855 [07:42<1:35:38,  2.10s/it, loss=1.2716, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 125/2855 [07:44<1:35:38,  2.10s/it, loss=1.1698, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 126/2855 [07:44<1:37:16,  2.14s/it, loss=1.1698, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 126/2855 [07:46<1:37:16,  2.14s/it, loss=1.0669, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 127/2855 [07:46<1:37:09,  2.14s/it, loss=1.0669, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 127/2855 [07:48<1:37:09,  2.14s/it, loss=0.9917, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 128/2855 [07:48<1:38:00,  2.16s/it, loss=0.9917, lr=3.00e-05]
Epoch 1/8:   4%|▍         | 128/2855 [07:50<1:38:00,  2.16s/it, loss=1.0604, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 129/2855 [07:50<1:37:15,  2.14s/it, loss=1.0604, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 129/2855 [07:52<1:37:15,  2.14s/it, loss=1.0461, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 130/2855 [07:52<1:36:44,  2.13s/it, loss=1.0461, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 130/2855 [07:55<1:36:44,  2.13s/it, loss=0.7517, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 131/2855 [07:55<1:35:54,  2.11s/it, loss=0.7517, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 131/2855 [07:57<1:35:54,  2.11s/it, loss=0.7053, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 132/2855 [07:57<1:36:59,  2.14s/it, loss=0.7053, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 132/2855 [07:59<1:36:59,  2.14s/it, loss=0.7128, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 133/2855 [07:59<1:37:15,  2.14s/it, loss=0.7128, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 133/2855 [08:01<1:37:15,  2.14s/it, loss=0.7827, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 134/2855 [08:01<1:37:45,  2.16s/it, loss=0.7827, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 134/2855 [08:03<1:37:45,  2.16s/it, loss=0.6872, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 135/2855 [08:03<1:36:53,  2.14s/it, loss=0.6872, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 135/2855 [08:05<1:36:53,  2.14s/it, loss=0.6611, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 136/2855 [08:05<1:36:40,  2.13s/it, loss=0.6611, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 136/2855 [08:08<1:36:40,  2.13s/it, loss=0.7728, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 137/2855 [08:08<1:38:28,  2.17s/it, loss=0.7728, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 137/2855 [08:10<1:38:28,  2.17s/it, loss=0.6535, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 138/2855 [08:10<1:40:31,  2.22s/it, loss=0.6535, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 138/2855 [08:13<1:40:31,  2.22s/it, loss=0.5926, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 139/2855 [08:13<1:49:33,  2.42s/it, loss=0.5926, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 139/2855 [08:21<1:49:33,  2.42s/it, loss=0.5872, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 140/2855 [08:21<3:02:50,  4.04s/it, loss=0.5872, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 140/2855 [08:23<3:02:50,  4.04s/it, loss=0.5954, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 141/2855 [08:23<2:40:20,  3.54s/it, loss=0.5954, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 141/2855 [08:26<2:40:20,  3.54s/it, loss=0.5457, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 142/2855 [08:26<2:29:29,  3.31s/it, loss=0.5457, lr=3.00e-05]
Epoch 1/8:   5%|▍         | 142/2855 [08:28<2:29:29,  3.31s/it, loss=0.4220, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 143/2855 [08:28<2:15:03,  2.99s/it, loss=0.4220, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 143/2855 [08:31<2:15:03,  2.99s/it, loss=0.4833, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 144/2855 [08:31<2:11:38,  2.91s/it, loss=0.4833, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 144/2855 [08:33<2:11:38,  2.91s/it, loss=0.4751, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 145/2855 [08:33<2:07:59,  2.83s/it, loss=0.4751, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 145/2855 [08:35<2:07:59,  2.83s/it, loss=0.4210, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 146/2855 [08:35<1:57:47,  2.61s/it, loss=0.4210, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 146/2855 [08:38<1:57:47,  2.61s/it, loss=0.6298, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 147/2855 [08:38<1:56:07,  2.57s/it, loss=0.6298, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 147/2855 [08:41<1:56:07,  2.57s/it, loss=0.2572, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 148/2855 [08:41<1:55:30,  2.56s/it, loss=0.2572, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 148/2855 [08:43<1:55:30,  2.56s/it, loss=0.5769, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 149/2855 [08:43<1:50:36,  2.45s/it, loss=0.5769, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 149/2855 [08:45<1:50:36,  2.45s/it, loss=0.4019, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 150/2855 [08:45<1:48:47,  2.41s/it, loss=0.4019, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 150/2855 [08:48<1:48:47,  2.41s/it, loss=0.4978, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 151/2855 [08:48<1:49:44,  2.44s/it, loss=0.4978, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 151/2855 [08:51<1:49:44,  2.44s/it, loss=0.3576, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 152/2855 [08:51<1:59:48,  2.66s/it, loss=0.3576, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 152/2855 [08:53<1:59:48,  2.66s/it, loss=0.4899, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 153/2855 [08:53<1:48:13,  2.40s/it, loss=0.4899, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 153/2855 [08:56<1:48:13,  2.40s/it, loss=0.3617, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 154/2855 [08:56<1:58:49,  2.64s/it, loss=0.3617, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 154/2855 [09:02<1:58:49,  2.64s/it, loss=0.3247, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 155/2855 [09:02<2:54:13,  3.87s/it, loss=0.3247, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 155/2855 [09:06<2:54:13,  3.87s/it, loss=0.4249, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 156/2855 [09:06<2:51:53,  3.82s/it, loss=0.4249, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 156/2855 [09:09<2:51:53,  3.82s/it, loss=0.2066, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 157/2855 [09:09<2:44:08,  3.65s/it, loss=0.2066, lr=3.00e-05]
Epoch 1/8:   5%|▌         | 157/2855 [09:12<2:44:08,  3.65s/it, loss=0.3366, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 158/2855 [09:12<2:30:18,  3.34s/it, loss=0.3366, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 158/2855 [09:14<2:30:18,  3.34s/it, loss=0.4669, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 159/2855 [09:14<2:12:30,  2.95s/it, loss=0.4669, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 159/2855 [09:16<2:12:30,  2.95s/it, loss=0.2509, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 160/2855 [09:16<2:00:40,  2.69s/it, loss=0.2509, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 160/2855 [09:18<2:00:40,  2.69s/it, loss=0.4357, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 161/2855 [09:18<1:53:22,  2.53s/it, loss=0.4357, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 161/2855 [09:20<1:53:22,  2.53s/it, loss=0.2148, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 162/2855 [09:20<1:48:00,  2.41s/it, loss=0.2148, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 162/2855 [09:22<1:48:00,  2.41s/it, loss=0.3768, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 163/2855 [09:22<1:42:57,  2.29s/it, loss=0.3768, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 163/2855 [09:25<1:42:57,  2.29s/it, loss=0.2714, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 164/2855 [09:25<1:40:29,  2.24s/it, loss=0.2714, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 164/2855 [09:27<1:40:29,  2.24s/it, loss=0.1968, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 165/2855 [09:27<1:39:35,  2.22s/it, loss=0.1968, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 165/2855 [09:29<1:39:35,  2.22s/it, loss=0.4452, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 166/2855 [09:29<1:38:57,  2.21s/it, loss=0.4452, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 166/2855 [09:31<1:38:57,  2.21s/it, loss=0.3703, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 167/2855 [09:31<1:37:09,  2.17s/it, loss=0.3703, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 167/2855 [09:33<1:37:09,  2.17s/it, loss=0.2015, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 168/2855 [09:33<1:34:50,  2.12s/it, loss=0.2015, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 168/2855 [09:35<1:34:50,  2.12s/it, loss=0.2205, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 169/2855 [09:35<1:33:16,  2.08s/it, loss=0.2205, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 169/2855 [09:37<1:33:16,  2.08s/it, loss=0.4451, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 170/2855 [09:37<1:31:03,  2.03s/it, loss=0.4451, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 170/2855 [09:39<1:31:03,  2.03s/it, loss=0.4195, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 171/2855 [09:39<1:32:07,  2.06s/it, loss=0.4195, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 171/2855 [09:41<1:32:07,  2.06s/it, loss=0.3899, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 172/2855 [09:41<1:31:52,  2.05s/it, loss=0.3899, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 172/2855 [09:43<1:31:52,  2.05s/it, loss=0.3092, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 173/2855 [09:43<1:31:06,  2.04s/it, loss=0.3092, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 173/2855 [09:45<1:31:06,  2.04s/it, loss=0.2670, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 174/2855 [09:45<1:30:28,  2.02s/it, loss=0.2670, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 174/2855 [09:47<1:30:28,  2.02s/it, loss=0.2613, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 175/2855 [09:47<1:28:33,  1.98s/it, loss=0.2613, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 175/2855 [09:50<1:28:33,  1.98s/it, loss=0.2636, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 176/2855 [09:50<1:44:37,  2.34s/it, loss=0.2636, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 176/2855 [09:54<1:44:37,  2.34s/it, loss=0.2656, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 177/2855 [09:54<2:06:38,  2.84s/it, loss=0.2656, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 177/2855 [09:56<2:06:38,  2.84s/it, loss=0.5725, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 178/2855 [09:56<1:59:21,  2.68s/it, loss=0.5725, lr=3.00e-05]
Epoch 1/8:   6%|▌         | 178/2855 [09:59<1:59:21,  2.68s/it, loss=0.2870, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 179/2855 [09:59<1:52:51,  2.53s/it, loss=0.2870, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 179/2855 [10:01<1:52:51,  2.53s/it, loss=0.2545, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 180/2855 [10:01<1:44:38,  2.35s/it, loss=0.2545, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 180/2855 [10:03<1:44:38,  2.35s/it, loss=0.3898, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 181/2855 [10:03<1:40:23,  2.25s/it, loss=0.3898, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 181/2855 [10:05<1:40:23,  2.25s/it, loss=0.0359, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 182/2855 [10:05<1:38:04,  2.20s/it, loss=0.0359, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 182/2855 [10:07<1:38:04,  2.20s/it, loss=0.2212, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 183/2855 [10:07<1:35:27,  2.14s/it, loss=0.2212, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 183/2855 [10:09<1:35:27,  2.14s/it, loss=0.2360, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 184/2855 [10:09<1:33:04,  2.09s/it, loss=0.2360, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 184/2855 [10:11<1:33:04,  2.09s/it, loss=0.3033, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 185/2855 [10:11<1:31:49,  2.06s/it, loss=0.3033, lr=3.00e-05]
Epoch 1/8:   6%|▋         | 185/2855 [10:13<1:31:49,  2.06s/it, loss=0.1077, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 186/2855 [10:13<1:31:44,  2.06s/it, loss=0.1077, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 186/2855 [10:15<1:31:44,  2.06s/it, loss=0.2036, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 187/2855 [10:15<1:32:14,  2.07s/it, loss=0.2036, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 187/2855 [10:17<1:32:14,  2.07s/it, loss=0.1772, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 188/2855 [10:17<1:32:14,  2.08s/it, loss=0.1772, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 188/2855 [10:19<1:32:14,  2.08s/it, loss=0.0646, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 189/2855 [10:19<1:32:41,  2.09s/it, loss=0.0646, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 189/2855 [10:21<1:32:41,  2.09s/it, loss=0.4670, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 190/2855 [10:21<1:32:54,  2.09s/it, loss=0.4670, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 190/2855 [10:23<1:32:54,  2.09s/it, loss=0.2411, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 191/2855 [10:23<1:31:28,  2.06s/it, loss=0.2411, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 191/2855 [10:25<1:31:28,  2.06s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 192/2855 [10:25<1:35:46,  2.16s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 192/2855 [10:28<1:35:46,  2.16s/it, loss=0.3243, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 193/2855 [10:28<1:35:49,  2.16s/it, loss=0.3243, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 193/2855 [10:30<1:35:49,  2.16s/it, loss=0.0849, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 194/2855 [10:30<1:43:33,  2.34s/it, loss=0.0849, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 194/2855 [10:32<1:43:33,  2.34s/it, loss=0.3714, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 195/2855 [10:32<1:37:25,  2.20s/it, loss=0.3714, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 195/2855 [10:34<1:37:25,  2.20s/it, loss=0.1828, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 196/2855 [10:34<1:36:42,  2.18s/it, loss=0.1828, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 196/2855 [10:36<1:36:42,  2.18s/it, loss=0.1109, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 197/2855 [10:36<1:35:31,  2.16s/it, loss=0.1109, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 197/2855 [10:38<1:35:31,  2.16s/it, loss=0.2237, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 198/2855 [10:38<1:33:35,  2.11s/it, loss=0.2237, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 198/2855 [10:41<1:33:35,  2.11s/it, loss=0.4015, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 199/2855 [10:41<1:34:25,  2.13s/it, loss=0.4015, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 199/2855 [10:43<1:34:25,  2.13s/it, loss=0.2646, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 200/2855 [10:43<1:35:02,  2.15s/it, loss=0.2646, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 200/2855 [10:45<1:35:02,  2.15s/it, loss=0.2603, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 201/2855 [10:45<1:34:44,  2.14s/it, loss=0.2603, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 201/2855 [10:47<1:34:44,  2.14s/it, loss=0.2414, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 202/2855 [10:47<1:31:50,  2.08s/it, loss=0.2414, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 202/2855 [10:49<1:31:50,  2.08s/it, loss=0.1920, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 203/2855 [10:49<1:30:01,  2.04s/it, loss=0.1920, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 203/2855 [10:51<1:30:01,  2.04s/it, loss=0.4247, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 204/2855 [10:51<1:26:24,  1.96s/it, loss=0.4247, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 204/2855 [10:53<1:26:24,  1.96s/it, loss=0.2283, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 205/2855 [10:53<1:26:51,  1.97s/it, loss=0.2283, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 205/2855 [10:55<1:26:51,  1.97s/it, loss=0.2851, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 206/2855 [10:55<1:27:11,  1.97s/it, loss=0.2851, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 206/2855 [10:56<1:27:11,  1.97s/it, loss=0.4119, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 207/2855 [10:56<1:26:03,  1.95s/it, loss=0.4119, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 207/2855 [10:58<1:26:03,  1.95s/it, loss=0.1070, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 208/2855 [10:58<1:25:16,  1.93s/it, loss=0.1070, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 208/2855 [11:00<1:25:16,  1.93s/it, loss=0.1473, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 209/2855 [11:00<1:24:59,  1.93s/it, loss=0.1473, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 209/2855 [11:02<1:24:59,  1.93s/it, loss=0.2483, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 210/2855 [11:02<1:24:42,  1.92s/it, loss=0.2483, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 210/2855 [11:06<1:24:42,  1.92s/it, loss=0.0716, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 211/2855 [11:06<1:52:38,  2.56s/it, loss=0.0716, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 211/2855 [11:08<1:52:38,  2.56s/it, loss=0.3339, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 212/2855 [11:08<1:43:18,  2.35s/it, loss=0.3339, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 212/2855 [11:11<1:43:18,  2.35s/it, loss=0.2176, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 213/2855 [11:11<1:49:20,  2.48s/it, loss=0.2176, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 213/2855 [11:13<1:49:20,  2.48s/it, loss=0.2886, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 214/2855 [11:13<1:39:02,  2.25s/it, loss=0.2886, lr=3.00e-05]
Epoch 1/8:   7%|▋         | 214/2855 [11:15<1:39:02,  2.25s/it, loss=0.1802, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 215/2855 [11:15<1:36:13,  2.19s/it, loss=0.1802, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 215/2855 [11:17<1:36:13,  2.19s/it, loss=0.2222, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 216/2855 [11:17<1:32:26,  2.10s/it, loss=0.2222, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 216/2855 [11:18<1:32:26,  2.10s/it, loss=0.2389, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 217/2855 [11:18<1:29:34,  2.04s/it, loss=0.2389, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 217/2855 [11:20<1:29:34,  2.04s/it, loss=0.1938, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 218/2855 [11:20<1:28:41,  2.02s/it, loss=0.1938, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 218/2855 [11:22<1:28:41,  2.02s/it, loss=0.1713, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 219/2855 [11:22<1:27:50,  2.00s/it, loss=0.1713, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 219/2855 [11:24<1:27:50,  2.00s/it, loss=0.0651, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 220/2855 [11:24<1:28:39,  2.02s/it, loss=0.0651, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 220/2855 [11:26<1:28:39,  2.02s/it, loss=0.2250, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 221/2855 [11:26<1:27:49,  2.00s/it, loss=0.2250, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 221/2855 [11:29<1:27:49,  2.00s/it, loss=0.2351, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 222/2855 [11:29<1:31:48,  2.09s/it, loss=0.2351, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 222/2855 [11:31<1:31:48,  2.09s/it, loss=0.1460, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 223/2855 [11:31<1:29:42,  2.04s/it, loss=0.1460, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 223/2855 [11:32<1:29:42,  2.04s/it, loss=0.3279, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 224/2855 [11:32<1:27:09,  1.99s/it, loss=0.3279, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 224/2855 [11:36<1:27:09,  1.99s/it, loss=0.1391, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 225/2855 [11:36<1:53:45,  2.60s/it, loss=0.1391, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 225/2855 [11:38<1:53:45,  2.60s/it, loss=0.3514, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 226/2855 [11:38<1:43:38,  2.37s/it, loss=0.3514, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 226/2855 [11:41<1:43:38,  2.37s/it, loss=0.2645, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 227/2855 [11:41<1:52:21,  2.57s/it, loss=0.2645, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 227/2855 [11:43<1:52:21,  2.57s/it, loss=0.2036, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 228/2855 [11:43<1:42:23,  2.34s/it, loss=0.2036, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 228/2855 [11:45<1:42:23,  2.34s/it, loss=0.1821, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 229/2855 [11:45<1:39:40,  2.28s/it, loss=0.1821, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 229/2855 [11:47<1:39:40,  2.28s/it, loss=0.1387, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 230/2855 [11:47<1:35:24,  2.18s/it, loss=0.1387, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 230/2855 [11:49<1:35:24,  2.18s/it, loss=0.3889, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 231/2855 [11:49<1:33:13,  2.13s/it, loss=0.3889, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 231/2855 [11:51<1:33:13,  2.13s/it, loss=0.3108, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 232/2855 [11:51<1:33:44,  2.14s/it, loss=0.3108, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 232/2855 [11:54<1:33:44,  2.14s/it, loss=0.2957, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 233/2855 [11:54<1:36:02,  2.20s/it, loss=0.2957, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 233/2855 [11:56<1:36:02,  2.20s/it, loss=0.4981, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 234/2855 [11:56<1:33:07,  2.13s/it, loss=0.4981, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 234/2855 [11:58<1:33:07,  2.13s/it, loss=0.2125, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 235/2855 [11:58<1:34:15,  2.16s/it, loss=0.2125, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 235/2855 [12:00<1:34:15,  2.16s/it, loss=0.4281, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 236/2855 [12:00<1:34:23,  2.16s/it, loss=0.4281, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 236/2855 [12:02<1:34:23,  2.16s/it, loss=0.5706, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 237/2855 [12:02<1:36:23,  2.21s/it, loss=0.5706, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 237/2855 [12:04<1:36:23,  2.21s/it, loss=0.4253, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 238/2855 [12:04<1:33:32,  2.14s/it, loss=0.4253, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 238/2855 [12:06<1:33:32,  2.14s/it, loss=0.1756, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 239/2855 [12:06<1:31:57,  2.11s/it, loss=0.1756, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 239/2855 [12:08<1:31:57,  2.11s/it, loss=0.1657, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 240/2855 [12:08<1:28:55,  2.04s/it, loss=0.1657, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 240/2855 [12:10<1:28:55,  2.04s/it, loss=0.0601, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 241/2855 [12:10<1:27:04,  2.00s/it, loss=0.0601, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 241/2855 [12:12<1:27:04,  2.00s/it, loss=0.1452, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 242/2855 [12:12<1:27:40,  2.01s/it, loss=0.1452, lr=3.00e-05]
Epoch 1/8:   8%|▊         | 242/2855 [12:14<1:27:40,  2.01s/it, loss=0.2947, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 243/2855 [12:14<1:27:43,  2.02s/it, loss=0.2947, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 243/2855 [12:16<1:27:43,  2.02s/it, loss=0.3738, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 244/2855 [12:16<1:26:27,  1.99s/it, loss=0.3738, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 244/2855 [12:18<1:26:27,  1.99s/it, loss=0.1562, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 245/2855 [12:18<1:25:57,  1.98s/it, loss=0.1562, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 245/2855 [12:20<1:25:57,  1.98s/it, loss=0.4589, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 246/2855 [12:20<1:23:49,  1.93s/it, loss=0.4589, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 246/2855 [12:22<1:23:49,  1.93s/it, loss=0.1500, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 247/2855 [12:22<1:24:22,  1.94s/it, loss=0.1500, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 247/2855 [12:24<1:24:22,  1.94s/it, loss=0.2182, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 248/2855 [12:24<1:24:35,  1.95s/it, loss=0.2182, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 248/2855 [12:26<1:24:35,  1.95s/it, loss=0.2965, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 249/2855 [12:26<1:23:46,  1.93s/it, loss=0.2965, lr=3.00e-05]
Epoch 1/8:   9%|▊         | 249/2855 [12:28<1:23:46,  1.93s/it, loss=0.2964, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 250/2855 [12:28<1:24:03,  1.94s/it, loss=0.2964, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 250/2855 [12:30<1:24:03,  1.94s/it, loss=0.2078, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 251/2855 [12:30<1:27:25,  2.01s/it, loss=0.2078, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 251/2855 [12:32<1:27:25,  2.01s/it, loss=0.2934, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 252/2855 [12:32<1:27:56,  2.03s/it, loss=0.2934, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 252/2855 [12:34<1:27:56,  2.03s/it, loss=0.4384, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 253/2855 [12:34<1:28:22,  2.04s/it, loss=0.4384, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 253/2855 [12:36<1:28:22,  2.04s/it, loss=0.0589, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 254/2855 [12:36<1:27:23,  2.02s/it, loss=0.0589, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 254/2855 [12:38<1:27:23,  2.02s/it, loss=0.1734, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 255/2855 [12:38<1:27:00,  2.01s/it, loss=0.1734, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 255/2855 [12:40<1:27:00,  2.01s/it, loss=0.3214, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 256/2855 [12:40<1:27:08,  2.01s/it, loss=0.3214, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 256/2855 [12:42<1:27:08,  2.01s/it, loss=0.2136, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 257/2855 [12:42<1:24:33,  1.95s/it, loss=0.2136, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 257/2855 [12:44<1:24:33,  1.95s/it, loss=0.0612, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 258/2855 [12:44<1:25:53,  1.98s/it, loss=0.0612, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 258/2855 [12:46<1:25:53,  1.98s/it, loss=0.2994, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 259/2855 [12:46<1:31:49,  2.12s/it, loss=0.2994, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 259/2855 [12:48<1:31:49,  2.12s/it, loss=0.3776, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 260/2855 [12:48<1:31:13,  2.11s/it, loss=0.3776, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 260/2855 [12:51<1:31:13,  2.11s/it, loss=0.1672, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 261/2855 [12:51<1:33:36,  2.17s/it, loss=0.1672, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 261/2855 [12:53<1:33:36,  2.17s/it, loss=0.3686, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 262/2855 [12:53<1:35:26,  2.21s/it, loss=0.3686, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 262/2855 [12:55<1:35:26,  2.21s/it, loss=0.1769, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 263/2855 [12:55<1:33:43,  2.17s/it, loss=0.1769, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 263/2855 [12:58<1:33:43,  2.17s/it, loss=0.3265, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 264/2855 [12:58<1:37:59,  2.27s/it, loss=0.3265, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 264/2855 [13:00<1:37:59,  2.27s/it, loss=0.1336, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 265/2855 [13:00<1:36:02,  2.23s/it, loss=0.1336, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 265/2855 [13:02<1:36:02,  2.23s/it, loss=0.4459, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 266/2855 [13:02<1:33:23,  2.16s/it, loss=0.4459, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 266/2855 [13:04<1:33:23,  2.16s/it, loss=0.2787, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 267/2855 [13:04<1:29:18,  2.07s/it, loss=0.2787, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 267/2855 [13:06<1:29:18,  2.07s/it, loss=0.2588, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 268/2855 [13:06<1:26:37,  2.01s/it, loss=0.2588, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 268/2855 [13:07<1:26:37,  2.01s/it, loss=0.1066, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 269/2855 [13:07<1:25:47,  1.99s/it, loss=0.1066, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 269/2855 [13:09<1:25:47,  1.99s/it, loss=0.1382, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 270/2855 [13:09<1:25:55,  1.99s/it, loss=0.1382, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 270/2855 [13:11<1:25:55,  1.99s/it, loss=0.1160, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 271/2855 [13:11<1:25:19,  1.98s/it, loss=0.1160, lr=3.00e-05]
Epoch 1/8:   9%|▉         | 271/2855 [13:14<1:25:19,  1.98s/it, loss=0.0962, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 272/2855 [13:14<1:27:55,  2.04s/it, loss=0.0962, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 272/2855 [13:16<1:27:55,  2.04s/it, loss=0.3379, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 273/2855 [13:16<1:30:19,  2.10s/it, loss=0.3379, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 273/2855 [13:18<1:30:19,  2.10s/it, loss=0.3552, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 274/2855 [13:18<1:32:40,  2.15s/it, loss=0.3552, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 274/2855 [13:20<1:32:40,  2.15s/it, loss=0.2450, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 275/2855 [13:20<1:33:02,  2.16s/it, loss=0.2450, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 275/2855 [13:22<1:33:02,  2.16s/it, loss=0.2122, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 276/2855 [13:22<1:32:36,  2.15s/it, loss=0.2122, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 276/2855 [13:25<1:32:36,  2.15s/it, loss=0.2340, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 277/2855 [13:25<1:32:13,  2.15s/it, loss=0.2340, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 277/2855 [13:27<1:32:13,  2.15s/it, loss=0.1259, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 278/2855 [13:27<1:30:05,  2.10s/it, loss=0.1259, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 278/2855 [13:29<1:30:05,  2.10s/it, loss=0.3441, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 279/2855 [13:29<1:28:49,  2.07s/it, loss=0.3441, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 279/2855 [13:31<1:28:49,  2.07s/it, loss=0.2747, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 280/2855 [13:31<1:27:24,  2.04s/it, loss=0.2747, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 280/2855 [13:33<1:27:24,  2.04s/it, loss=0.1574, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 281/2855 [13:33<1:26:58,  2.03s/it, loss=0.1574, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 281/2855 [13:34<1:26:58,  2.03s/it, loss=0.3095, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 282/2855 [13:34<1:25:52,  2.00s/it, loss=0.3095, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 282/2855 [13:36<1:25:52,  2.00s/it, loss=0.3831, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 283/2855 [13:36<1:25:49,  2.00s/it, loss=0.3831, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 283/2855 [13:38<1:25:49,  2.00s/it, loss=0.3746, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 284/2855 [13:38<1:25:39,  2.00s/it, loss=0.3746, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 284/2855 [13:40<1:25:39,  2.00s/it, loss=0.0125, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 285/2855 [13:40<1:25:31,  2.00s/it, loss=0.0125, lr=3.00e-05]
Epoch 1/8:  10%|▉         | 285/2855 [13:42<1:25:31,  2.00s/it, loss=0.1280, lr=3.00e-05]
Epoch 1/8:  10%|█         | 286/2855 [13:42<1:24:35,  1.98s/it, loss=0.1280, lr=3.00e-05]
Epoch 1/8:  10%|█         | 286/2855 [13:44<1:24:35,  1.98s/it, loss=0.5512, lr=3.00e-05]
Epoch 1/8:  10%|█         | 287/2855 [13:44<1:23:14,  1.94s/it, loss=0.5512, lr=3.00e-05]
Epoch 1/8:  10%|█         | 287/2855 [13:46<1:23:14,  1.94s/it, loss=0.1768, lr=3.00e-05]
Epoch 1/8:  10%|█         | 288/2855 [13:46<1:23:21,  1.95s/it, loss=0.1768, lr=3.00e-05]
Epoch 1/8:  10%|█         | 288/2855 [13:48<1:23:21,  1.95s/it, loss=0.2092, lr=3.00e-05]
Epoch 1/8:  10%|█         | 289/2855 [13:48<1:21:22,  1.90s/it, loss=0.2092, lr=3.00e-05]
Epoch 1/8:  10%|█         | 289/2855 [13:50<1:21:22,  1.90s/it, loss=0.0976, lr=3.00e-05]
Epoch 1/8:  10%|█         | 290/2855 [13:50<1:21:00,  1.89s/it, loss=0.0976, lr=3.00e-05]
Epoch 1/8:  10%|█         | 290/2855 [13:52<1:21:00,  1.89s/it, loss=0.2768, lr=3.00e-05]
Epoch 1/8:  10%|█         | 291/2855 [13:52<1:21:48,  1.91s/it, loss=0.2768, lr=3.00e-05]
Epoch 1/8:  10%|█         | 291/2855 [13:54<1:21:48,  1.91s/it, loss=0.1106, lr=3.00e-05]
Epoch 1/8:  10%|█         | 292/2855 [13:54<1:19:30,  1.86s/it, loss=0.1106, lr=3.00e-05]
Epoch 1/8:  10%|█         | 292/2855 [13:56<1:19:30,  1.86s/it, loss=0.2235, lr=3.00e-05]
Epoch 1/8:  10%|█         | 293/2855 [13:56<1:21:21,  1.91s/it, loss=0.2235, lr=3.00e-05]
Epoch 1/8:  10%|█         | 293/2855 [13:58<1:21:21,  1.91s/it, loss=0.1977, lr=3.00e-05]
Epoch 1/8:  10%|█         | 294/2855 [13:58<1:21:48,  1.92s/it, loss=0.1977, lr=3.00e-05]
Epoch 1/8:  10%|█         | 294/2855 [13:59<1:21:48,  1.92s/it, loss=0.1508, lr=3.00e-05]
Epoch 1/8:  10%|█         | 295/2855 [13:59<1:21:44,  1.92s/it, loss=0.1508, lr=3.00e-05]
Epoch 1/8:  10%|█         | 295/2855 [14:02<1:21:44,  1.92s/it, loss=0.1121, lr=3.00e-05]
Epoch 1/8:  10%|█         | 296/2855 [14:02<1:24:20,  1.98s/it, loss=0.1121, lr=3.00e-05]
Epoch 1/8:  10%|█         | 296/2855 [14:04<1:24:20,  1.98s/it, loss=0.3037, lr=3.00e-05]
Epoch 1/8:  10%|█         | 297/2855 [14:04<1:25:40,  2.01s/it, loss=0.3037, lr=3.00e-05]
Epoch 1/8:  10%|█         | 297/2855 [14:06<1:25:40,  2.01s/it, loss=0.3547, lr=3.00e-05]
Epoch 1/8:  10%|█         | 298/2855 [14:06<1:26:14,  2.02s/it, loss=0.3547, lr=3.00e-05]
Epoch 1/8:  10%|█         | 298/2855 [14:08<1:26:14,  2.02s/it, loss=0.1636, lr=3.00e-05]
Epoch 1/8:  10%|█         | 299/2855 [14:08<1:25:20,  2.00s/it, loss=0.1636, lr=3.00e-05]
Epoch 1/8:  10%|█         | 299/2855 [14:10<1:25:20,  2.00s/it, loss=0.1848, lr=3.00e-05]
Epoch 1/8:  11%|█         | 300/2855 [14:10<1:26:30,  2.03s/it, loss=0.1848, lr=3.00e-05]
Epoch 1/8:  11%|█         | 300/2855 [14:12<1:26:30,  2.03s/it, loss=0.3174, lr=3.00e-05]
Epoch 1/8:  11%|█         | 301/2855 [14:12<1:25:14,  2.00s/it, loss=0.3174, lr=3.00e-05]
Epoch 1/8:  11%|█         | 301/2855 [14:14<1:25:14,  2.00s/it, loss=0.4356, lr=3.00e-05]
Epoch 1/8:  11%|█         | 302/2855 [14:14<1:25:33,  2.01s/it, loss=0.4356, lr=3.00e-05]
Epoch 1/8:  11%|█         | 302/2855 [14:16<1:25:33,  2.01s/it, loss=0.1063, lr=3.00e-05]
Epoch 1/8:  11%|█         | 303/2855 [14:16<1:24:16,  1.98s/it, loss=0.1063, lr=3.00e-05]
Epoch 1/8:  11%|█         | 303/2855 [14:18<1:24:16,  1.98s/it, loss=0.1687, lr=3.00e-05]
Epoch 1/8:  11%|█         | 304/2855 [14:18<1:24:14,  1.98s/it, loss=0.1687, lr=3.00e-05]
Epoch 1/8:  11%|█         | 304/2855 [14:20<1:24:14,  1.98s/it, loss=0.1923, lr=3.00e-05]
Epoch 1/8:  11%|█         | 305/2855 [14:20<1:27:24,  2.06s/it, loss=0.1923, lr=3.00e-05]
Epoch 1/8:  11%|█         | 305/2855 [14:22<1:27:24,  2.06s/it, loss=0.2485, lr=3.00e-05]
Epoch 1/8:  11%|█         | 306/2855 [14:22<1:26:27,  2.04s/it, loss=0.2485, lr=3.00e-05]
Epoch 1/8:  11%|█         | 306/2855 [14:24<1:26:27,  2.04s/it, loss=0.0735, lr=3.00e-05]
Epoch 1/8:  11%|█         | 307/2855 [14:24<1:26:49,  2.04s/it, loss=0.0735, lr=3.00e-05]
Epoch 1/8:  11%|█         | 307/2855 [14:26<1:26:49,  2.04s/it, loss=0.2565, lr=3.00e-05]
Epoch 1/8:  11%|█         | 308/2855 [14:26<1:25:25,  2.01s/it, loss=0.2565, lr=3.00e-05]
Epoch 1/8:  11%|█         | 308/2855 [14:28<1:25:25,  2.01s/it, loss=0.2210, lr=3.00e-05]
Epoch 1/8:  11%|█         | 309/2855 [14:28<1:28:01,  2.07s/it, loss=0.2210, lr=3.00e-05]
Epoch 1/8:  11%|█         | 309/2855 [14:30<1:28:01,  2.07s/it, loss=0.1208, lr=3.00e-05]
Epoch 1/8:  11%|█         | 310/2855 [14:30<1:28:16,  2.08s/it, loss=0.1208, lr=3.00e-05]
Epoch 1/8:  11%|█         | 310/2855 [14:32<1:28:16,  2.08s/it, loss=0.2406, lr=3.00e-05]
Epoch 1/8:  11%|█         | 311/2855 [14:32<1:26:28,  2.04s/it, loss=0.2406, lr=3.00e-05]
Epoch 1/8:  11%|█         | 311/2855 [14:34<1:26:28,  2.04s/it, loss=0.1849, lr=3.00e-05]
Epoch 1/8:  11%|█         | 312/2855 [14:34<1:26:50,  2.05s/it, loss=0.1849, lr=3.00e-05]
Epoch 1/8:  11%|█         | 312/2855 [14:36<1:26:50,  2.05s/it, loss=0.3118, lr=3.00e-05]
Epoch 1/8:  11%|█         | 313/2855 [14:36<1:25:41,  2.02s/it, loss=0.3118, lr=3.00e-05]
Epoch 1/8:  11%|█         | 313/2855 [14:38<1:25:41,  2.02s/it, loss=0.3039, lr=3.00e-05]
Epoch 1/8:  11%|█         | 314/2855 [14:38<1:25:20,  2.02s/it, loss=0.3039, lr=3.00e-05]
Epoch 1/8:  11%|█         | 314/2855 [14:40<1:25:20,  2.02s/it, loss=0.0815, lr=3.00e-05]
Epoch 1/8:  11%|█         | 315/2855 [14:40<1:23:34,  1.97s/it, loss=0.0815, lr=3.00e-05]
Epoch 1/8:  11%|█         | 315/2855 [14:42<1:23:34,  1.97s/it, loss=0.0958, lr=3.00e-05]
Epoch 1/8:  11%|█         | 316/2855 [14:42<1:22:50,  1.96s/it, loss=0.0958, lr=3.00e-05]
Epoch 1/8:  11%|█         | 316/2855 [14:44<1:22:50,  1.96s/it, loss=0.0357, lr=3.00e-05]
Epoch 1/8:  11%|█         | 317/2855 [14:44<1:21:41,  1.93s/it, loss=0.0357, lr=3.00e-05]
Epoch 1/8:  11%|█         | 317/2855 [14:46<1:21:41,  1.93s/it, loss=0.3808, lr=3.00e-05]
Epoch 1/8:  11%|█         | 318/2855 [14:46<1:19:57,  1.89s/it, loss=0.3808, lr=3.00e-05]
Epoch 1/8:  11%|█         | 318/2855 [15:10<1:19:57,  1.89s/it, loss=0.2748, lr=3.00e-05]
Epoch 1/8:  11%|█         | 319/2855 [15:10<6:02:58,  8.59s/it, loss=0.2748, lr=3.00e-05]
Epoch 1/8:  11%|█         | 319/2855 [15:12<6:02:58,  8.59s/it, loss=0.2578, lr=3.00e-05]
Epoch 1/8:  11%|█         | 320/2855 [15:12<4:40:45,  6.65s/it, loss=0.2578, lr=3.00e-05]
Epoch 1/8:  11%|█         | 320/2855 [15:14<4:40:45,  6.65s/it, loss=0.3066, lr=3.00e-05]
Epoch 1/8:  11%|█         | 321/2855 [15:14<3:43:53,  5.30s/it, loss=0.3066, lr=3.00e-05]
Epoch 1/8:  11%|█         | 321/2855 [15:16<3:43:53,  5.30s/it, loss=0.2974, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 322/2855 [15:16<3:06:56,  4.43s/it, loss=0.2974, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 322/2855 [15:18<3:06:56,  4.43s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 323/2855 [15:18<2:36:19,  3.70s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 323/2855 [15:20<2:36:19,  3.70s/it, loss=0.1931, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 324/2855 [15:20<2:11:55,  3.13s/it, loss=0.1931, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 324/2855 [15:22<2:11:55,  3.13s/it, loss=0.1453, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 325/2855 [15:22<1:58:19,  2.81s/it, loss=0.1453, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 325/2855 [15:25<1:58:19,  2.81s/it, loss=0.0512, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 326/2855 [15:25<1:51:19,  2.64s/it, loss=0.0512, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 326/2855 [15:27<1:51:19,  2.64s/it, loss=0.0949, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 327/2855 [15:27<1:43:17,  2.45s/it, loss=0.0949, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 327/2855 [15:29<1:43:17,  2.45s/it, loss=0.1959, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 328/2855 [15:29<1:38:19,  2.33s/it, loss=0.1959, lr=3.00e-05]
Epoch 1/8:  11%|█▏        | 328/2855 [15:31<1:38:19,  2.33s/it, loss=0.2608, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 329/2855 [15:31<1:34:41,  2.25s/it, loss=0.2608, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 329/2855 [15:33<1:34:41,  2.25s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 330/2855 [15:33<1:31:38,  2.18s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 330/2855 [15:35<1:31:38,  2.18s/it, loss=0.3488, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 331/2855 [15:35<1:29:43,  2.13s/it, loss=0.3488, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 331/2855 [15:37<1:29:43,  2.13s/it, loss=0.3787, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 332/2855 [15:37<1:27:58,  2.09s/it, loss=0.3787, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 332/2855 [15:39<1:27:58,  2.09s/it, loss=0.0706, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 333/2855 [15:39<1:27:14,  2.08s/it, loss=0.0706, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 333/2855 [15:41<1:27:14,  2.08s/it, loss=0.2048, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 334/2855 [15:41<1:25:56,  2.05s/it, loss=0.2048, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 334/2855 [15:43<1:25:56,  2.05s/it, loss=0.1286, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 335/2855 [15:43<1:25:25,  2.03s/it, loss=0.1286, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 335/2855 [15:45<1:25:25,  2.03s/it, loss=0.1232, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 336/2855 [15:45<1:25:40,  2.04s/it, loss=0.1232, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 336/2855 [15:47<1:25:40,  2.04s/it, loss=0.1781, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 337/2855 [15:47<1:24:38,  2.02s/it, loss=0.1781, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 337/2855 [15:49<1:24:38,  2.02s/it, loss=0.1659, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 338/2855 [15:49<1:24:16,  2.01s/it, loss=0.1659, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 338/2855 [15:51<1:24:16,  2.01s/it, loss=0.1664, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 339/2855 [15:51<1:23:35,  1.99s/it, loss=0.1664, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 339/2855 [15:53<1:23:35,  1.99s/it, loss=0.2197, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 340/2855 [15:53<1:24:36,  2.02s/it, loss=0.2197, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 340/2855 [15:55<1:24:36,  2.02s/it, loss=0.0676, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 341/2855 [15:55<1:25:31,  2.04s/it, loss=0.0676, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 341/2855 [15:57<1:25:31,  2.04s/it, loss=0.2552, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 342/2855 [15:57<1:24:26,  2.02s/it, loss=0.2552, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 342/2855 [15:59<1:24:26,  2.02s/it, loss=0.1584, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 343/2855 [15:59<1:24:24,  2.02s/it, loss=0.1584, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 343/2855 [16:01<1:24:24,  2.02s/it, loss=0.2865, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 344/2855 [16:01<1:24:23,  2.02s/it, loss=0.2865, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 344/2855 [16:03<1:24:23,  2.02s/it, loss=0.1919, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 345/2855 [16:03<1:24:41,  2.02s/it, loss=0.1919, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 345/2855 [16:05<1:24:41,  2.02s/it, loss=0.1960, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 346/2855 [16:05<1:25:54,  2.05s/it, loss=0.1960, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 346/2855 [16:07<1:25:54,  2.05s/it, loss=0.2280, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 347/2855 [16:07<1:25:21,  2.04s/it, loss=0.2280, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 347/2855 [16:09<1:25:21,  2.04s/it, loss=0.3084, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 348/2855 [16:09<1:25:45,  2.05s/it, loss=0.3084, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 348/2855 [16:11<1:25:45,  2.05s/it, loss=0.5327, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 349/2855 [16:11<1:25:47,  2.05s/it, loss=0.5327, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 349/2855 [16:13<1:25:47,  2.05s/it, loss=0.4068, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 350/2855 [16:13<1:25:51,  2.06s/it, loss=0.4068, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 350/2855 [16:15<1:25:51,  2.06s/it, loss=0.0979, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 351/2855 [16:15<1:25:18,  2.04s/it, loss=0.0979, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 351/2855 [16:17<1:25:18,  2.04s/it, loss=0.3019, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 352/2855 [16:17<1:24:49,  2.03s/it, loss=0.3019, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 352/2855 [16:19<1:24:49,  2.03s/it, loss=0.1576, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 353/2855 [16:19<1:24:48,  2.03s/it, loss=0.1576, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 353/2855 [16:21<1:24:48,  2.03s/it, loss=0.1887, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 354/2855 [16:21<1:23:40,  2.01s/it, loss=0.1887, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 354/2855 [16:23<1:23:40,  2.01s/it, loss=0.1493, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 355/2855 [16:23<1:24:36,  2.03s/it, loss=0.1493, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 355/2855 [16:25<1:24:36,  2.03s/it, loss=0.1687, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 356/2855 [16:25<1:25:34,  2.05s/it, loss=0.1687, lr=3.00e-05]
Epoch 1/8:  12%|█▏        | 356/2855 [16:27<1:25:34,  2.05s/it, loss=0.2773, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 357/2855 [16:27<1:25:20,  2.05s/it, loss=0.2773, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 357/2855 [16:30<1:25:20,  2.05s/it, loss=0.2387, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 358/2855 [16:30<1:25:24,  2.05s/it, loss=0.2387, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 358/2855 [16:34<1:25:24,  2.05s/it, loss=0.2453, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 359/2855 [16:34<1:53:12,  2.72s/it, loss=0.2453, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 359/2855 [16:36<1:53:12,  2.72s/it, loss=0.2227, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 360/2855 [16:36<1:45:18,  2.53s/it, loss=0.2227, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 360/2855 [16:38<1:45:18,  2.53s/it, loss=0.3914, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 361/2855 [16:38<1:42:16,  2.46s/it, loss=0.3914, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 361/2855 [16:40<1:42:16,  2.46s/it, loss=0.2176, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 362/2855 [16:40<1:36:32,  2.32s/it, loss=0.2176, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 362/2855 [16:42<1:36:32,  2.32s/it, loss=0.1297, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 363/2855 [16:42<1:33:38,  2.25s/it, loss=0.1297, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 363/2855 [16:44<1:33:38,  2.25s/it, loss=0.2187, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 364/2855 [16:44<1:30:40,  2.18s/it, loss=0.2187, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 364/2855 [16:46<1:30:40,  2.18s/it, loss=0.2816, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 365/2855 [16:46<1:28:17,  2.13s/it, loss=0.2816, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 365/2855 [16:48<1:28:17,  2.13s/it, loss=0.5060, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 366/2855 [16:48<1:26:49,  2.09s/it, loss=0.5060, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 366/2855 [16:50<1:26:49,  2.09s/it, loss=0.1167, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 367/2855 [16:50<1:25:59,  2.07s/it, loss=0.1167, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 367/2855 [16:52<1:25:59,  2.07s/it, loss=0.0290, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 368/2855 [16:52<1:23:58,  2.03s/it, loss=0.0290, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 368/2855 [16:54<1:23:58,  2.03s/it, loss=0.2462, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 369/2855 [16:54<1:24:39,  2.04s/it, loss=0.2462, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 369/2855 [16:56<1:24:39,  2.04s/it, loss=0.3384, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 370/2855 [16:56<1:23:11,  2.01s/it, loss=0.3384, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 370/2855 [16:58<1:23:11,  2.01s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 371/2855 [16:58<1:23:29,  2.02s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 371/2855 [17:01<1:23:29,  2.02s/it, loss=0.2406, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 372/2855 [17:01<1:26:33,  2.09s/it, loss=0.2406, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 372/2855 [17:03<1:26:33,  2.09s/it, loss=0.0534, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 373/2855 [17:03<1:25:35,  2.07s/it, loss=0.0534, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 373/2855 [17:05<1:25:35,  2.07s/it, loss=0.2617, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 374/2855 [17:05<1:26:58,  2.10s/it, loss=0.2617, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 374/2855 [17:07<1:26:58,  2.10s/it, loss=0.0027, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 375/2855 [17:07<1:27:53,  2.13s/it, loss=0.0027, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 375/2855 [17:09<1:27:53,  2.13s/it, loss=0.1730, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 376/2855 [17:09<1:28:49,  2.15s/it, loss=0.1730, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 376/2855 [17:11<1:28:49,  2.15s/it, loss=0.0809, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 377/2855 [17:11<1:26:19,  2.09s/it, loss=0.0809, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 377/2855 [17:13<1:26:19,  2.09s/it, loss=0.1818, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 378/2855 [17:13<1:25:17,  2.07s/it, loss=0.1818, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 378/2855 [17:15<1:25:17,  2.07s/it, loss=0.1011, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 379/2855 [17:15<1:26:10,  2.09s/it, loss=0.1011, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 379/2855 [17:17<1:26:10,  2.09s/it, loss=0.2673, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 380/2855 [17:17<1:26:27,  2.10s/it, loss=0.2673, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 380/2855 [17:19<1:26:27,  2.10s/it, loss=0.2823, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 381/2855 [17:19<1:26:11,  2.09s/it, loss=0.2823, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 381/2855 [17:22<1:26:11,  2.09s/it, loss=0.1355, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 382/2855 [17:22<1:26:42,  2.10s/it, loss=0.1355, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 382/2855 [17:24<1:26:42,  2.10s/it, loss=0.2618, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 383/2855 [17:24<1:26:28,  2.10s/it, loss=0.2618, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 383/2855 [17:26<1:26:28,  2.10s/it, loss=0.3380, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 384/2855 [17:26<1:25:29,  2.08s/it, loss=0.3380, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 384/2855 [17:28<1:25:29,  2.08s/it, loss=0.2171, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 385/2855 [17:28<1:24:53,  2.06s/it, loss=0.2171, lr=3.00e-05]
Epoch 1/8:  13%|█▎        | 385/2855 [17:30<1:24:53,  2.06s/it, loss=0.2945, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 386/2855 [17:30<1:27:16,  2.12s/it, loss=0.2945, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 386/2855 [17:32<1:27:16,  2.12s/it, loss=0.2065, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 387/2855 [17:32<1:26:45,  2.11s/it, loss=0.2065, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 387/2855 [17:34<1:26:45,  2.11s/it, loss=0.2723, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 388/2855 [17:34<1:25:49,  2.09s/it, loss=0.2723, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 388/2855 [17:36<1:25:49,  2.09s/it, loss=0.4055, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 389/2855 [17:36<1:25:44,  2.09s/it, loss=0.4055, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 389/2855 [17:38<1:25:44,  2.09s/it, loss=0.1119, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 390/2855 [17:38<1:24:37,  2.06s/it, loss=0.1119, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 390/2855 [17:40<1:24:37,  2.06s/it, loss=0.1671, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 391/2855 [17:40<1:24:39,  2.06s/it, loss=0.1671, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 391/2855 [17:42<1:24:39,  2.06s/it, loss=0.0486, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 392/2855 [17:42<1:25:00,  2.07s/it, loss=0.0486, lr=3.00e-05]
Epoch 1/8:  14%|█▎        | 392/2855 [17:44<1:25:00,  2.07s/it, loss=0.1849, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 393/2855 [17:44<1:24:12,  2.05s/it, loss=0.1849, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 393/2855 [17:46<1:24:12,  2.05s/it, loss=0.3101, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 394/2855 [17:46<1:23:28,  2.04s/it, loss=0.3101, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 394/2855 [17:48<1:23:28,  2.04s/it, loss=0.2616, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 395/2855 [17:48<1:22:49,  2.02s/it, loss=0.2616, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 395/2855 [17:50<1:22:49,  2.02s/it, loss=0.2929, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 396/2855 [17:50<1:23:12,  2.03s/it, loss=0.2929, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 396/2855 [17:52<1:23:12,  2.03s/it, loss=0.2674, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 397/2855 [17:52<1:22:24,  2.01s/it, loss=0.2674, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 397/2855 [17:54<1:22:24,  2.01s/it, loss=0.3297, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 398/2855 [17:54<1:23:25,  2.04s/it, loss=0.3297, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 398/2855 [18:00<1:23:25,  2.04s/it, loss=0.3863, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 399/2855 [18:00<2:11:33,  3.21s/it, loss=0.3863, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 399/2855 [18:03<2:11:33,  3.21s/it, loss=0.3281, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 400/2855 [18:03<2:00:32,  2.95s/it, loss=0.3281, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 400/2855 [18:05<2:00:32,  2.95s/it, loss=0.1377, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 401/2855 [18:05<1:51:21,  2.72s/it, loss=0.1377, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 401/2855 [18:07<1:51:21,  2.72s/it, loss=0.3162, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 402/2855 [18:07<1:42:42,  2.51s/it, loss=0.3162, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 402/2855 [18:09<1:42:42,  2.51s/it, loss=0.1191, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 403/2855 [18:09<1:37:08,  2.38s/it, loss=0.1191, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 403/2855 [18:11<1:37:08,  2.38s/it, loss=0.1644, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 404/2855 [18:11<1:34:58,  2.32s/it, loss=0.1644, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 404/2855 [18:13<1:34:58,  2.32s/it, loss=0.2748, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 405/2855 [18:13<1:33:33,  2.29s/it, loss=0.2748, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 405/2855 [18:16<1:33:33,  2.29s/it, loss=0.3902, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 406/2855 [18:16<1:32:50,  2.27s/it, loss=0.3902, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 406/2855 [18:18<1:32:50,  2.27s/it, loss=0.4044, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 407/2855 [18:18<1:31:40,  2.25s/it, loss=0.4044, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 407/2855 [18:21<1:31:40,  2.25s/it, loss=0.1775, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 408/2855 [18:21<1:41:29,  2.49s/it, loss=0.1775, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 408/2855 [18:24<1:41:29,  2.49s/it, loss=0.1577, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 409/2855 [18:24<1:43:32,  2.54s/it, loss=0.1577, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 409/2855 [18:26<1:43:32,  2.54s/it, loss=0.2702, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 410/2855 [18:26<1:44:11,  2.56s/it, loss=0.2702, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 410/2855 [18:28<1:44:11,  2.56s/it, loss=0.3159, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 411/2855 [18:28<1:40:09,  2.46s/it, loss=0.3159, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 411/2855 [18:31<1:40:09,  2.46s/it, loss=0.1495, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 412/2855 [18:31<1:36:48,  2.38s/it, loss=0.1495, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 412/2855 [18:33<1:36:48,  2.38s/it, loss=0.3308, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 413/2855 [18:33<1:37:43,  2.40s/it, loss=0.3308, lr=3.00e-05]
Epoch 1/8:  14%|█▍        | 413/2855 [18:35<1:37:43,  2.40s/it, loss=0.1836, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 414/2855 [18:35<1:35:02,  2.34s/it, loss=0.1836, lr=3.00e-05]
Epoch 1/8:  15%|��▍        | 414/2855 [18:37<1:35:02,  2.34s/it, loss=0.3299, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 415/2855 [18:37<1:31:37,  2.25s/it, loss=0.3299, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 415/2855 [18:40<1:31:37,  2.25s/it, loss=0.1853, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 416/2855 [18:40<1:32:06,  2.27s/it, loss=0.1853, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 416/2855 [18:42<1:32:06,  2.27s/it, loss=0.2352, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 417/2855 [18:42<1:28:44,  2.18s/it, loss=0.2352, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 417/2855 [18:43<1:28:44,  2.18s/it, loss=0.1270, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 418/2855 [18:43<1:25:13,  2.10s/it, loss=0.1270, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 418/2855 [18:45<1:25:13,  2.10s/it, loss=0.2821, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 419/2855 [18:45<1:22:41,  2.04s/it, loss=0.2821, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 419/2855 [18:47<1:22:41,  2.04s/it, loss=0.4008, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 420/2855 [18:47<1:20:42,  1.99s/it, loss=0.4008, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 420/2855 [18:49<1:20:42,  1.99s/it, loss=0.1638, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 421/2855 [18:49<1:20:47,  1.99s/it, loss=0.1638, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 421/2855 [18:51<1:20:47,  1.99s/it, loss=0.1418, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 422/2855 [18:51<1:21:26,  2.01s/it, loss=0.1418, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 422/2855 [18:53<1:21:26,  2.01s/it, loss=0.3163, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 423/2855 [18:53<1:21:57,  2.02s/it, loss=0.3163, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 423/2855 [18:55<1:21:57,  2.02s/it, loss=0.0776, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 424/2855 [18:55<1:23:15,  2.05s/it, loss=0.0776, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 424/2855 [18:57<1:23:15,  2.05s/it, loss=0.1159, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 425/2855 [18:57<1:21:50,  2.02s/it, loss=0.1159, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 425/2855 [19:00<1:21:50,  2.02s/it, loss=0.1546, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 426/2855 [19:00<1:22:30,  2.04s/it, loss=0.1546, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 426/2855 [19:02<1:22:30,  2.04s/it, loss=0.1936, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 427/2855 [19:02<1:22:12,  2.03s/it, loss=0.1936, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 427/2855 [19:04<1:22:12,  2.03s/it, loss=0.2728, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 428/2855 [19:04<1:21:39,  2.02s/it, loss=0.2728, lr=3.00e-05]
Epoch 1/8:  15%|█▍        | 428/2855 [19:06<1:21:39,  2.02s/it, loss=0.0650, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 429/2855 [19:06<1:21:24,  2.01s/it, loss=0.0650, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 429/2855 [19:08<1:21:24,  2.01s/it, loss=0.2661, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 430/2855 [19:08<1:22:07,  2.03s/it, loss=0.2661, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 430/2855 [19:10<1:22:07,  2.03s/it, loss=0.3898, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 431/2855 [19:10<1:25:25,  2.11s/it, loss=0.3898, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 431/2855 [19:12<1:25:25,  2.11s/it, loss=0.2964, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 432/2855 [19:12<1:27:52,  2.18s/it, loss=0.2964, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 432/2855 [19:14<1:27:52,  2.18s/it, loss=0.1248, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 433/2855 [19:14<1:26:47,  2.15s/it, loss=0.1248, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 433/2855 [19:16<1:26:47,  2.15s/it, loss=0.1335, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 434/2855 [19:16<1:24:21,  2.09s/it, loss=0.1335, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 434/2855 [19:18<1:24:21,  2.09s/it, loss=0.2301, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 435/2855 [19:18<1:23:39,  2.07s/it, loss=0.2301, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 435/2855 [19:20<1:23:39,  2.07s/it, loss=0.1839, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 436/2855 [19:20<1:22:55,  2.06s/it, loss=0.1839, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 436/2855 [19:22<1:22:55,  2.06s/it, loss=0.0827, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 437/2855 [19:22<1:22:24,  2.04s/it, loss=0.0827, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 437/2855 [19:28<1:22:24,  2.04s/it, loss=0.2038, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 438/2855 [19:28<2:11:02,  3.25s/it, loss=0.2038, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 438/2855 [19:30<2:11:02,  3.25s/it, loss=0.1568, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 439/2855 [19:30<1:52:19,  2.79s/it, loss=0.1568, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 439/2855 [19:32<1:52:19,  2.79s/it, loss=0.0763, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 440/2855 [19:32<1:46:59,  2.66s/it, loss=0.0763, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 440/2855 [19:34<1:46:59,  2.66s/it, loss=0.0983, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 441/2855 [19:34<1:35:06,  2.36s/it, loss=0.0983, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 441/2855 [19:36<1:35:06,  2.36s/it, loss=0.2360, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 442/2855 [19:36<1:28:01,  2.19s/it, loss=0.2360, lr=3.00e-05]
Epoch 1/8:  15%|█▌        | 442/2855 [19:38<1:28:01,  2.19s/it, loss=0.1325, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 443/2855 [19:38<1:22:42,  2.06s/it, loss=0.1325, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 443/2855 [19:40<1:22:42,  2.06s/it, loss=0.1176, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 444/2855 [19:40<1:20:12,  2.00s/it, loss=0.1176, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 444/2855 [19:41<1:20:12,  2.00s/it, loss=0.1421, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 445/2855 [19:41<1:17:56,  1.94s/it, loss=0.1421, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 445/2855 [19:43<1:17:56,  1.94s/it, loss=0.1667, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 446/2855 [19:43<1:18:11,  1.95s/it, loss=0.1667, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 446/2855 [19:45<1:18:11,  1.95s/it, loss=0.2883, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 447/2855 [19:45<1:20:01,  1.99s/it, loss=0.2883, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 447/2855 [19:47<1:20:01,  1.99s/it, loss=0.4092, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 448/2855 [19:47<1:20:57,  2.02s/it, loss=0.4092, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 448/2855 [19:49<1:20:57,  2.02s/it, loss=0.1766, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 449/2855 [19:49<1:20:40,  2.01s/it, loss=0.1766, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 449/2855 [19:51<1:20:40,  2.01s/it, loss=0.0999, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 450/2855 [19:51<1:20:47,  2.02s/it, loss=0.0999, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 450/2855 [19:54<1:20:47,  2.02s/it, loss=0.2298, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 451/2855 [19:54<1:20:50,  2.02s/it, loss=0.2298, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 451/2855 [19:55<1:20:50,  2.02s/it, loss=0.1789, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 452/2855 [19:55<1:20:07,  2.00s/it, loss=0.1789, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 452/2855 [19:57<1:20:07,  2.00s/it, loss=0.4385, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 453/2855 [19:57<1:20:17,  2.01s/it, loss=0.4385, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 453/2855 [20:00<1:20:17,  2.01s/it, loss=0.4304, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 454/2855 [20:00<1:20:13,  2.00s/it, loss=0.4304, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 454/2855 [20:01<1:20:13,  2.00s/it, loss=0.0743, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 455/2855 [20:01<1:19:55,  2.00s/it, loss=0.0743, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 455/2855 [20:04<1:19:55,  2.00s/it, loss=0.2388, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 456/2855 [20:04<1:20:56,  2.02s/it, loss=0.2388, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 456/2855 [20:06<1:20:56,  2.02s/it, loss=0.2520, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 457/2855 [20:06<1:21:05,  2.03s/it, loss=0.2520, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 457/2855 [20:08<1:21:05,  2.03s/it, loss=0.2653, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 458/2855 [20:08<1:20:31,  2.02s/it, loss=0.2653, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 458/2855 [20:10<1:20:31,  2.02s/it, loss=0.1539, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 459/2855 [20:10<1:19:53,  2.00s/it, loss=0.1539, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 459/2855 [20:12<1:19:53,  2.00s/it, loss=0.0818, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 460/2855 [20:12<1:21:04,  2.03s/it, loss=0.0818, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 460/2855 [20:14<1:21:04,  2.03s/it, loss=0.2359, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 461/2855 [20:14<1:19:41,  2.00s/it, loss=0.2359, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 461/2855 [20:16<1:19:41,  2.00s/it, loss=0.2348, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 462/2855 [20:16<1:20:06,  2.01s/it, loss=0.2348, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 462/2855 [20:18<1:20:06,  2.01s/it, loss=0.3159, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 463/2855 [20:18<1:19:40,  2.00s/it, loss=0.3159, lr=3.00e-05]
Epoch 1/8:  16%|█▌        | 463/2855 [20:20<1:19:40,  2.00s/it, loss=0.2063, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 464/2855 [20:20<1:18:54,  1.98s/it, loss=0.2063, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 464/2855 [20:22<1:18:54,  1.98s/it, loss=0.0935, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 465/2855 [20:22<1:19:48,  2.00s/it, loss=0.0935, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 465/2855 [20:24<1:19:48,  2.00s/it, loss=0.0273, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 466/2855 [20:24<1:19:52,  2.01s/it, loss=0.0273, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 466/2855 [20:26<1:19:52,  2.01s/it, loss=0.2278, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 467/2855 [20:26<1:19:33,  2.00s/it, loss=0.2278, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 467/2855 [20:27<1:19:33,  2.00s/it, loss=0.1686, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 468/2855 [20:27<1:18:32,  1.97s/it, loss=0.1686, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 468/2855 [20:30<1:18:32,  1.97s/it, loss=0.1104, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 469/2855 [20:30<1:19:21,  2.00s/it, loss=0.1104, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 469/2855 [20:32<1:19:21,  2.00s/it, loss=0.1975, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 470/2855 [20:32<1:19:46,  2.01s/it, loss=0.1975, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 470/2855 [20:34<1:19:46,  2.01s/it, loss=0.0591, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 471/2855 [20:34<1:19:49,  2.01s/it, loss=0.0591, lr=3.00e-05]
Epoch 1/8:  16%|█▋        | 471/2855 [20:36<1:19:49,  2.01s/it, loss=0.3365, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 472/2855 [20:36<1:19:25,  2.00s/it, loss=0.3365, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 472/2855 [20:38<1:19:25,  2.00s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 473/2855 [20:38<1:19:08,  1.99s/it, loss=0.3134, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 473/2855 [20:39<1:19:08,  1.99s/it, loss=0.1453, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 474/2855 [20:39<1:18:02,  1.97s/it, loss=0.1453, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 474/2855 [20:42<1:18:02,  1.97s/it, loss=0.0592, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 475/2855 [20:42<1:19:43,  2.01s/it, loss=0.0592, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 475/2855 [20:45<1:19:43,  2.01s/it, loss=0.2286, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 476/2855 [20:45<1:34:04,  2.37s/it, loss=0.2286, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 476/2855 [20:51<1:34:04,  2.37s/it, loss=0.1035, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 477/2855 [20:51<2:16:33,  3.45s/it, loss=0.1035, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 477/2855 [20:58<2:16:33,  3.45s/it, loss=0.0448, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 478/2855 [20:58<2:58:40,  4.51s/it, loss=0.0448, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 478/2855 [21:01<2:58:40,  4.51s/it, loss=0.1283, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 479/2855 [21:01<2:43:15,  4.12s/it, loss=0.1283, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 479/2855 [21:06<2:43:15,  4.12s/it, loss=0.0981, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 480/2855 [21:06<2:56:47,  4.47s/it, loss=0.0981, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 480/2855 [21:09<2:56:47,  4.47s/it, loss=0.3340, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 481/2855 [21:09<2:34:54,  3.92s/it, loss=0.3340, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 481/2855 [21:11<2:34:54,  3.92s/it, loss=0.2303, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 482/2855 [21:11<2:10:38,  3.30s/it, loss=0.2303, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 482/2855 [21:13<2:10:38,  3.30s/it, loss=0.1703, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 483/2855 [21:13<1:54:56,  2.91s/it, loss=0.1703, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 483/2855 [21:15<1:54:56,  2.91s/it, loss=0.2885, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 484/2855 [21:15<1:43:27,  2.62s/it, loss=0.2885, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 484/2855 [21:17<1:43:27,  2.62s/it, loss=0.1177, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 485/2855 [21:17<1:34:52,  2.40s/it, loss=0.1177, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 485/2855 [21:19<1:34:52,  2.40s/it, loss=0.0821, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 486/2855 [21:19<1:35:36,  2.42s/it, loss=0.0821, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 486/2855 [21:21<1:35:36,  2.42s/it, loss=0.1692, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 487/2855 [21:21<1:31:52,  2.33s/it, loss=0.1692, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 487/2855 [21:23<1:31:52,  2.33s/it, loss=0.3528, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 488/2855 [21:23<1:27:57,  2.23s/it, loss=0.3528, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 488/2855 [21:25<1:27:57,  2.23s/it, loss=0.2011, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 489/2855 [21:25<1:25:10,  2.16s/it, loss=0.2011, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 489/2855 [21:27<1:25:10,  2.16s/it, loss=0.1058, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 490/2855 [21:27<1:21:57,  2.08s/it, loss=0.1058, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 490/2855 [21:29<1:21:57,  2.08s/it, loss=0.0880, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 491/2855 [21:29<1:22:49,  2.10s/it, loss=0.0880, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 491/2855 [21:31<1:22:49,  2.10s/it, loss=0.4435, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 492/2855 [21:31<1:23:00,  2.11s/it, loss=0.4435, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 492/2855 [21:33<1:23:00,  2.11s/it, loss=0.2474, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 493/2855 [21:33<1:22:31,  2.10s/it, loss=0.2474, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 493/2855 [21:35<1:22:31,  2.10s/it, loss=0.1829, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 494/2855 [21:35<1:22:23,  2.09s/it, loss=0.1829, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 494/2855 [21:38<1:22:23,  2.09s/it, loss=0.2390, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 495/2855 [21:38<1:22:23,  2.09s/it, loss=0.2390, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 495/2855 [21:40<1:22:23,  2.09s/it, loss=0.1975, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 496/2855 [21:40<1:22:39,  2.10s/it, loss=0.1975, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 496/2855 [21:42<1:22:39,  2.10s/it, loss=0.2887, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 497/2855 [21:42<1:22:42,  2.10s/it, loss=0.2887, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 497/2855 [21:44<1:22:42,  2.10s/it, loss=0.1380, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 498/2855 [21:44<1:24:41,  2.16s/it, loss=0.1380, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 498/2855 [21:50<1:24:41,  2.16s/it, loss=0.2446, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 499/2855 [21:50<2:12:14,  3.37s/it, loss=0.2446, lr=3.00e-05]
Epoch 1/8:  17%|█▋        | 499/2855 [21:53<2:12:14,  3.37s/it, loss=0.2100, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 500/2855 [21:53<2:04:20,  3.17s/it, loss=0.2100, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 500/2855 [21:55<2:04:20,  3.17s/it, loss=0.2507, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 501/2855 [21:55<1:55:36,  2.95s/it, loss=0.2507, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 501/2855 [21:58<1:55:36,  2.95s/it, loss=0.0938, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 502/2855 [21:58<1:48:40,  2.77s/it, loss=0.0938, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 502/2855 [22:00<1:48:40,  2.77s/it, loss=0.2534, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 503/2855 [22:00<1:47:53,  2.75s/it, loss=0.2534, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 503/2855 [22:02<1:47:53,  2.75s/it, loss=0.1320, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 504/2855 [22:02<1:39:12,  2.53s/it, loss=0.1320, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 504/2855 [22:05<1:39:12,  2.53s/it, loss=0.2749, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 505/2855 [22:05<1:35:19,  2.43s/it, loss=0.2749, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 505/2855 [22:07<1:35:19,  2.43s/it, loss=0.1312, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 506/2855 [22:07<1:32:11,  2.36s/it, loss=0.1312, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 506/2855 [22:09<1:32:11,  2.36s/it, loss=0.2034, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 507/2855 [22:09<1:31:32,  2.34s/it, loss=0.2034, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 507/2855 [22:12<1:31:32,  2.34s/it, loss=0.3281, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 508/2855 [22:12<1:32:52,  2.37s/it, loss=0.3281, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 508/2855 [22:14<1:32:52,  2.37s/it, loss=0.2837, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 509/2855 [22:14<1:32:57,  2.38s/it, loss=0.2837, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 509/2855 [22:16<1:32:57,  2.38s/it, loss=0.1953, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 510/2855 [22:16<1:31:49,  2.35s/it, loss=0.1953, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 510/2855 [22:18<1:31:49,  2.35s/it, loss=0.3413, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 511/2855 [22:18<1:30:01,  2.30s/it, loss=0.3413, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 511/2855 [22:21<1:30:01,  2.30s/it, loss=0.1354, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 512/2855 [22:21<1:31:41,  2.35s/it, loss=0.1354, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 512/2855 [22:23<1:31:41,  2.35s/it, loss=0.2080, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 513/2855 [22:23<1:28:53,  2.28s/it, loss=0.2080, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 513/2855 [22:25<1:28:53,  2.28s/it, loss=0.1371, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 514/2855 [22:25<1:29:41,  2.30s/it, loss=0.1371, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 514/2855 [22:28<1:29:41,  2.30s/it, loss=0.2692, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 515/2855 [22:28<1:28:31,  2.27s/it, loss=0.2692, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 515/2855 [22:30<1:28:31,  2.27s/it, loss=0.1393, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 516/2855 [22:30<1:27:30,  2.24s/it, loss=0.1393, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 516/2855 [22:36<1:27:30,  2.24s/it, loss=0.2857, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 517/2855 [22:36<2:19:41,  3.59s/it, loss=0.2857, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 517/2855 [22:38<2:19:41,  3.59s/it, loss=0.1516, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 518/2855 [22:38<1:53:18,  2.91s/it, loss=0.1516, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 518/2855 [22:41<1:53:18,  2.91s/it, loss=0.1785, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 519/2855 [22:41<1:51:45,  2.87s/it, loss=0.1785, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 519/2855 [22:43<1:51:45,  2.87s/it, loss=0.2286, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 520/2855 [22:43<1:40:46,  2.59s/it, loss=0.2286, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 520/2855 [22:45<1:40:46,  2.59s/it, loss=0.2899, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 521/2855 [22:45<1:36:51,  2.49s/it, loss=0.2899, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 521/2855 [22:47<1:36:51,  2.49s/it, loss=0.0440, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 522/2855 [22:47<1:34:50,  2.44s/it, loss=0.0440, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 522/2855 [22:49<1:34:50,  2.44s/it, loss=0.2099, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 523/2855 [22:49<1:30:57,  2.34s/it, loss=0.2099, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 523/2855 [22:51<1:30:57,  2.34s/it, loss=0.0103, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 524/2855 [22:51<1:29:05,  2.29s/it, loss=0.0103, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 524/2855 [22:54<1:29:05,  2.29s/it, loss=0.1317, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 525/2855 [22:54<1:27:07,  2.24s/it, loss=0.1317, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 525/2855 [22:56<1:27:07,  2.24s/it, loss=0.3131, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 526/2855 [22:56<1:26:01,  2.22s/it, loss=0.3131, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 526/2855 [22:58<1:26:01,  2.22s/it, loss=0.1923, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 527/2855 [22:58<1:24:24,  2.18s/it, loss=0.1923, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 527/2855 [23:00<1:24:24,  2.18s/it, loss=0.0994, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 528/2855 [23:00<1:23:08,  2.14s/it, loss=0.0994, lr=3.00e-05]
Epoch 1/8:  18%|█▊        | 528/2855 [23:02<1:23:08,  2.14s/it, loss=0.1555, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 529/2855 [23:02<1:22:08,  2.12s/it, loss=0.1555, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 529/2855 [23:04<1:22:08,  2.12s/it, loss=0.1117, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 530/2855 [23:04<1:20:17,  2.07s/it, loss=0.1117, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 530/2855 [23:06<1:20:17,  2.07s/it, loss=0.2160, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 531/2855 [23:06<1:19:42,  2.06s/it, loss=0.2160, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 531/2855 [23:08<1:19:42,  2.06s/it, loss=0.2322, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 532/2855 [23:08<1:17:44,  2.01s/it, loss=0.2322, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 532/2855 [23:10<1:17:44,  2.01s/it, loss=0.1953, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 533/2855 [23:10<1:15:33,  1.95s/it, loss=0.1953, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 533/2855 [23:12<1:15:33,  1.95s/it, loss=0.2784, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 534/2855 [23:12<1:16:08,  1.97s/it, loss=0.2784, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 534/2855 [23:13<1:16:08,  1.97s/it, loss=0.1683, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 535/2855 [23:13<1:13:58,  1.91s/it, loss=0.1683, lr=3.00e-05]
Epoch 1/8:  19%|█▊        | 535/2855 [23:15<1:13:58,  1.91s/it, loss=0.1546, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 536/2855 [23:15<1:12:34,  1.88s/it, loss=0.1546, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 536/2855 [23:17<1:12:34,  1.88s/it, loss=0.1542, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 537/2855 [23:17<1:14:11,  1.92s/it, loss=0.1542, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 537/2855 [23:19<1:14:11,  1.92s/it, loss=0.2509, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 538/2855 [23:19<1:12:01,  1.87s/it, loss=0.2509, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 538/2855 [23:21<1:12:01,  1.87s/it, loss=0.3151, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 539/2855 [23:21<1:12:10,  1.87s/it, loss=0.3151, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 539/2855 [23:23<1:12:10,  1.87s/it, loss=0.2033, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 540/2855 [23:23<1:18:31,  2.04s/it, loss=0.2033, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 540/2855 [23:25<1:18:31,  2.04s/it, loss=0.3790, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 541/2855 [23:25<1:19:40,  2.07s/it, loss=0.3790, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 541/2855 [23:27<1:19:40,  2.07s/it, loss=0.2822, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 542/2855 [23:27<1:19:33,  2.06s/it, loss=0.2822, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 542/2855 [23:29<1:19:33,  2.06s/it, loss=0.3477, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 543/2855 [23:29<1:18:53,  2.05s/it, loss=0.3477, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 543/2855 [23:32<1:18:53,  2.05s/it, loss=0.0776, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 544/2855 [23:32<1:22:43,  2.15s/it, loss=0.0776, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 544/2855 [23:34<1:22:43,  2.15s/it, loss=0.2325, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 545/2855 [23:34<1:20:52,  2.10s/it, loss=0.2325, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 545/2855 [23:36<1:20:52,  2.10s/it, loss=0.3833, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 546/2855 [23:36<1:19:34,  2.07s/it, loss=0.3833, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 546/2855 [23:38<1:19:34,  2.07s/it, loss=0.3127, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 547/2855 [23:38<1:23:13,  2.16s/it, loss=0.3127, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 547/2855 [23:40<1:23:13,  2.16s/it, loss=0.2572, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 548/2855 [23:40<1:21:26,  2.12s/it, loss=0.2572, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 548/2855 [23:43<1:21:26,  2.12s/it, loss=0.4287, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 549/2855 [23:43<1:23:56,  2.18s/it, loss=0.4287, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 549/2855 [23:45<1:23:56,  2.18s/it, loss=0.1048, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 550/2855 [23:45<1:26:36,  2.25s/it, loss=0.1048, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 550/2855 [23:47<1:26:36,  2.25s/it, loss=0.1743, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 551/2855 [23:47<1:27:24,  2.28s/it, loss=0.1743, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 551/2855 [23:49<1:27:24,  2.28s/it, loss=0.1260, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 552/2855 [23:49<1:25:52,  2.24s/it, loss=0.1260, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 552/2855 [23:52<1:25:52,  2.24s/it, loss=0.1963, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 553/2855 [23:52<1:25:38,  2.23s/it, loss=0.1963, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 553/2855 [23:54<1:25:38,  2.23s/it, loss=0.1049, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 554/2855 [23:54<1:25:17,  2.22s/it, loss=0.1049, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 554/2855 [23:56<1:25:17,  2.22s/it, loss=0.3895, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 555/2855 [23:56<1:24:07,  2.19s/it, loss=0.3895, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 555/2855 [23:58<1:24:07,  2.19s/it, loss=0.0957, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 556/2855 [23:58<1:20:58,  2.11s/it, loss=0.0957, lr=3.00e-05]
Epoch 1/8:  19%|█▉        | 556/2855 [24:05<1:20:58,  2.11s/it, loss=0.1671, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 557/2855 [24:05<2:16:08,  3.55s/it, loss=0.1671, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 557/2855 [24:06<2:16:08,  3.55s/it, loss=0.1403, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 558/2855 [24:06<1:53:49,  2.97s/it, loss=0.1403, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 558/2855 [24:09<1:53:49,  2.97s/it, loss=0.2085, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 559/2855 [24:09<1:51:09,  2.90s/it, loss=0.2085, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 559/2855 [24:11<1:51:09,  2.90s/it, loss=0.0493, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 560/2855 [24:11<1:39:40,  2.61s/it, loss=0.0493, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 560/2855 [24:13<1:39:40,  2.61s/it, loss=0.1499, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 561/2855 [24:13<1:29:59,  2.35s/it, loss=0.1499, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 561/2855 [24:15<1:29:59,  2.35s/it, loss=0.1759, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 562/2855 [24:15<1:26:47,  2.27s/it, loss=0.1759, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 562/2855 [24:17<1:26:47,  2.27s/it, loss=0.1735, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 563/2855 [24:17<1:23:22,  2.18s/it, loss=0.1735, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 563/2855 [24:19<1:23:22,  2.18s/it, loss=0.2845, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 564/2855 [24:19<1:17:50,  2.04s/it, loss=0.2845, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 564/2855 [24:21<1:17:50,  2.04s/it, loss=0.1212, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 565/2855 [24:21<1:17:24,  2.03s/it, loss=0.1212, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 565/2855 [24:23<1:17:24,  2.03s/it, loss=0.0838, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 566/2855 [24:23<1:17:48,  2.04s/it, loss=0.0838, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 566/2855 [24:25<1:17:48,  2.04s/it, loss=0.2960, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 567/2855 [24:25<1:18:54,  2.07s/it, loss=0.2960, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 567/2855 [24:27<1:18:54,  2.07s/it, loss=0.1280, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 568/2855 [24:27<1:18:48,  2.07s/it, loss=0.1280, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 568/2855 [24:29<1:18:48,  2.07s/it, loss=0.2884, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 569/2855 [24:29<1:19:33,  2.09s/it, loss=0.2884, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 569/2855 [24:31<1:19:33,  2.09s/it, loss=0.2626, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 570/2855 [24:31<1:19:14,  2.08s/it, loss=0.2626, lr=3.00e-05]
Epoch 1/8:  20%|█▉        | 570/2855 [24:33<1:19:14,  2.08s/it, loss=0.3233, lr=3.00e-05]
Epoch 1/8:  20%|██        | 571/2855 [24:33<1:19:03,  2.08s/it, loss=0.3233, lr=3.00e-05]
Epoch 1/8:  20%|██        | 571/2855 [24:35<1:19:03,  2.08s/it, loss=0.0953, lr=3.00e-05]
Epoch 1/8:  20%|██        | 572/2855 [24:35<1:21:25,  2.14s/it, loss=0.0953, lr=3.00e-05]
Epoch 1/8:  20%|██        | 572/2855 [24:38<1:21:25,  2.14s/it, loss=0.1449, lr=3.00e-05]
Epoch 1/8:  20%|██        | 573/2855 [24:38<1:22:38,  2.17s/it, loss=0.1449, lr=3.00e-05]
Epoch 1/8:  20%|██        | 573/2855 [24:40<1:22:38,  2.17s/it, loss=0.2686, lr=3.00e-05]
Epoch 1/8:  20%|██        | 574/2855 [24:40<1:27:36,  2.30s/it, loss=0.2686, lr=3.00e-05]
Epoch 1/8:  20%|██        | 574/2855 [24:45<1:27:36,  2.30s/it, loss=0.3862, lr=3.00e-05]
Epoch 1/8:  20%|██        | 575/2855 [24:45<1:57:43,  3.10s/it, loss=0.3862, lr=3.00e-05]
Epoch 1/8:  20%|██        | 575/2855 [24:48<1:57:43,  3.10s/it, loss=0.3594, lr=3.00e-05]
Epoch 1/8:  20%|██        | 576/2855 [24:48<1:54:16,  3.01s/it, loss=0.3594, lr=3.00e-05]
Epoch 1/8:  20%|██        | 576/2855 [24:50<1:54:16,  3.01s/it, loss=0.3044, lr=3.00e-05]
Epoch 1/8:  20%|██        | 577/2855 [24:50<1:46:26,  2.80s/it, loss=0.3044, lr=3.00e-05]
Epoch 1/8:  20%|██        | 577/2855 [24:53<1:46:26,  2.80s/it, loss=0.3486, lr=3.00e-05]
Epoch 1/8:  20%|██        | 578/2855 [24:53<1:40:03,  2.64s/it, loss=0.3486, lr=3.00e-05]
Epoch 1/8:  20%|██        | 578/2855 [24:55<1:40:03,  2.64s/it, loss=0.1838, lr=3.00e-05]
Epoch 1/8:  20%|██        | 579/2855 [24:55<1:35:11,  2.51s/it, loss=0.1838, lr=3.00e-05]
Epoch 1/8:  20%|██        | 579/2855 [24:57<1:35:11,  2.51s/it, loss=0.1564, lr=3.00e-05]
Epoch 1/8:  20%|██        | 580/2855 [24:57<1:30:19,  2.38s/it, loss=0.1564, lr=3.00e-05]
Epoch 1/8:  20%|██        | 580/2855 [24:59<1:30:19,  2.38s/it, loss=0.1372, lr=3.00e-05]
Epoch 1/8:  20%|██        | 581/2855 [24:59<1:28:06,  2.32s/it, loss=0.1372, lr=3.00e-05]
Epoch 1/8:  20%|██        | 581/2855 [25:01<1:28:06,  2.32s/it, loss=0.1143, lr=3.00e-05]
Epoch 1/8:  20%|██        | 582/2855 [25:01<1:25:11,  2.25s/it, loss=0.1143, lr=3.00e-05]
Epoch 1/8:  20%|██        | 582/2855 [25:03<1:25:11,  2.25s/it, loss=0.1858, lr=3.00e-05]
Epoch 1/8:  20%|██        | 583/2855 [25:03<1:23:14,  2.20s/it, loss=0.1858, lr=3.00e-05]
Epoch 1/8:  20%|██        | 583/2855 [25:05<1:23:14,  2.20s/it, loss=0.4128, lr=3.00e-05]
Epoch 1/8:  20%|██        | 584/2855 [25:05<1:21:55,  2.16s/it, loss=0.4128, lr=3.00e-05]
Epoch 1/8:  20%|██        | 584/2855 [25:08<1:21:55,  2.16s/it, loss=0.2189, lr=3.00e-05]
Epoch 1/8:  20%|██        | 585/2855 [25:08<1:27:49,  2.32s/it, loss=0.2189, lr=3.00e-05]
Epoch 1/8:  20%|██        | 585/2855 [25:10<1:27:49,  2.32s/it, loss=0.2392, lr=3.00e-05]
Epoch 1/8:  21%|██        | 586/2855 [25:10<1:27:08,  2.30s/it, loss=0.2392, lr=3.00e-05]
Epoch 1/8:  21%|██        | 586/2855 [25:13<1:27:08,  2.30s/it, loss=0.1543, lr=3.00e-05]
Epoch 1/8:  21%|██        | 587/2855 [25:13<1:29:35,  2.37s/it, loss=0.1543, lr=3.00e-05]
Epoch 1/8:  21%|██        | 587/2855 [25:15<1:29:35,  2.37s/it, loss=0.0657, lr=3.00e-05]
Epoch 1/8:  21%|██        | 588/2855 [25:15<1:31:01,  2.41s/it, loss=0.0657, lr=3.00e-05]
Epoch 1/8:  21%|██        | 588/2855 [25:18<1:31:01,  2.41s/it, loss=0.0342, lr=3.00e-05]
Epoch 1/8:  21%|██        | 589/2855 [25:18<1:28:37,  2.35s/it, loss=0.0342, lr=3.00e-05]
Epoch 1/8:  21%|██        | 589/2855 [25:20<1:28:37,  2.35s/it, loss=0.1787, lr=3.00e-05]
Epoch 1/8:  21%|██        | 590/2855 [25:20<1:26:17,  2.29s/it, loss=0.1787, lr=3.00e-05]
Epoch 1/8:  21%|██        | 590/2855 [25:22<1:26:17,  2.29s/it, loss=0.1961, lr=3.00e-05]
Epoch 1/8:  21%|██        | 591/2855 [25:22<1:24:58,  2.25s/it, loss=0.1961, lr=3.00e-05]
Epoch 1/8:  21%|██        | 591/2855 [25:24<1:24:58,  2.25s/it, loss=0.2630, lr=3.00e-05]
Epoch 1/8:  21%|██        | 592/2855 [25:24<1:23:46,  2.22s/it, loss=0.2630, lr=3.00e-05]
Epoch 1/8:  21%|██        | 592/2855 [25:26<1:23:46,  2.22s/it, loss=0.1641, lr=3.00e-05]
Epoch 1/8:  21%|██        | 593/2855 [25:26<1:22:51,  2.20s/it, loss=0.1641, lr=3.00e-05]
Epoch 1/8:  21%|██        | 593/2855 [25:28<1:22:51,  2.20s/it, loss=0.4343, lr=3.00e-05]
Epoch 1/8:  21%|██        | 594/2855 [25:28<1:21:39,  2.17s/it, loss=0.4343, lr=3.00e-05]
Epoch 1/8:  21%|██        | 594/2855 [25:30<1:21:39,  2.17s/it, loss=0.1899, lr=3.00e-05]
Epoch 1/8:  21%|██        | 595/2855 [25:30<1:22:09,  2.18s/it, loss=0.1899, lr=3.00e-05]
Epoch 1/8:  21%|██        | 595/2855 [25:36<1:22:09,  2.18s/it, loss=0.3846, lr=3.00e-05]
Epoch 1/8:  21%|██        | 596/2855 [25:36<2:05:11,  3.33s/it, loss=0.3846, lr=3.00e-05]
Epoch 1/8:  21%|██        | 596/2855 [25:38<2:05:11,  3.33s/it, loss=0.3505, lr=3.00e-05]
Epoch 1/8:  21%|██        | 597/2855 [25:38<1:43:23,  2.75s/it, loss=0.3505, lr=3.00e-05]
Epoch 1/8:  21%|██        | 597/2855 [25:41<1:43:23,  2.75s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:  21%|██        | 598/2855 [25:41<1:43:55,  2.76s/it, loss=0.2026, lr=3.00e-05]
Epoch 1/8:  21%|██        | 598/2855 [25:43<1:43:55,  2.76s/it, loss=0.4022, lr=3.00e-05]
Epoch 1/8:  21%|██        | 599/2855 [25:43<1:41:04,  2.69s/it, loss=0.4022, lr=3.00e-05]
Epoch 1/8:  21%|██        | 599/2855 [25:46<1:41:04,  2.69s/it, loss=0.2419, lr=3.00e-05]
Epoch 1/8:  21%|██        | 600/2855 [25:46<1:42:53,  2.74s/it, loss=0.2419, lr=3.00e-05]
Epoch 1/8:  21%|██        | 600/2855 [25:48<1:42:53,  2.74s/it, loss=0.3338, lr=3.00e-05]
Epoch 1/8:  21%|██        | 601/2855 [25:48<1:37:54,  2.61s/it, loss=0.3338, lr=3.00e-05]
Epoch 1/8:  21%|██        | 601/2855 [25:51<1:37:54,  2.61s/it, loss=0.0857, lr=3.00e-05]
Epoch 1/8:  21%|██        | 602/2855 [25:51<1:35:54,  2.55s/it, loss=0.0857, lr=3.00e-05]
Epoch 1/8:  21%|██        | 602/2855 [25:53<1:35:54,  2.55s/it, loss=0.2491, lr=3.00e-05]
Epoch 1/8:  21%|██        | 603/2855 [25:53<1:33:47,  2.50s/it, loss=0.2491, lr=3.00e-05]
Epoch 1/8:  21%|██        | 603/2855 [25:55<1:33:47,  2.50s/it, loss=0.0628, lr=2.99e-05]
Epoch 1/8:  21%|██        | 604/2855 [25:55<1:30:04,  2.40s/it, loss=0.0628, lr=2.99e-05]
Epoch 1/8:  21%|██        | 604/2855 [25:57<1:30:04,  2.40s/it, loss=0.2304, lr=2.99e-05]
Epoch 1/8:  21%|██        | 605/2855 [25:57<1:27:00,  2.32s/it, loss=0.2304, lr=2.99e-05]
Epoch 1/8:  21%|██        | 605/2855 [26:00<1:27:00,  2.32s/it, loss=0.1274, lr=2.99e-05]
Epoch 1/8:  21%|██        | 606/2855 [26:00<1:25:43,  2.29s/it, loss=0.1274, lr=2.99e-05]
Epoch 1/8:  21%|██        | 606/2855 [26:02<1:25:43,  2.29s/it, loss=0.1083, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 607/2855 [26:02<1:24:32,  2.26s/it, loss=0.1083, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 607/2855 [26:04<1:24:32,  2.26s/it, loss=0.2335, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 608/2855 [26:04<1:23:00,  2.22s/it, loss=0.2335, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 608/2855 [26:06<1:23:00,  2.22s/it, loss=0.0557, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 609/2855 [26:06<1:22:34,  2.21s/it, loss=0.0557, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 609/2855 [26:08<1:22:34,  2.21s/it, loss=0.4083, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 610/2855 [26:08<1:21:27,  2.18s/it, loss=0.4083, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 610/2855 [26:10<1:21:27,  2.18s/it, loss=0.2625, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 611/2855 [26:10<1:21:05,  2.17s/it, loss=0.2625, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 611/2855 [26:13<1:21:05,  2.17s/it, loss=0.1713, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 612/2855 [26:13<1:21:19,  2.18s/it, loss=0.1713, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 612/2855 [26:15<1:21:19,  2.18s/it, loss=0.0397, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 613/2855 [26:15<1:20:40,  2.16s/it, loss=0.0397, lr=2.99e-05]
Epoch 1/8:  21%|██▏       | 613/2855 [26:17<1:20:40,  2.16s/it, loss=0.0615, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 614/2855 [26:17<1:19:29,  2.13s/it, loss=0.0615, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 614/2855 [26:19<1:19:29,  2.13s/it, loss=0.2810, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 615/2855 [26:19<1:19:47,  2.14s/it, loss=0.2810, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 615/2855 [26:21<1:19:47,  2.14s/it, loss=0.3604, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 616/2855 [26:21<1:20:04,  2.15s/it, loss=0.3604, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 616/2855 [26:23<1:20:04,  2.15s/it, loss=0.2791, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 617/2855 [26:23<1:20:41,  2.16s/it, loss=0.2791, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 617/2855 [26:26<1:20:41,  2.16s/it, loss=0.1501, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 618/2855 [26:26<1:24:03,  2.25s/it, loss=0.1501, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 618/2855 [26:28<1:24:03,  2.25s/it, loss=0.1119, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 619/2855 [26:28<1:24:23,  2.26s/it, loss=0.1119, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 619/2855 [26:30<1:24:23,  2.26s/it, loss=0.1806, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 620/2855 [26:30<1:23:38,  2.25s/it, loss=0.1806, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 620/2855 [26:32<1:23:38,  2.25s/it, loss=0.2661, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 621/2855 [26:32<1:22:13,  2.21s/it, loss=0.2661, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 621/2855 [26:35<1:22:13,  2.21s/it, loss=0.1469, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 622/2855 [26:35<1:23:18,  2.24s/it, loss=0.1469, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 622/2855 [26:37<1:23:18,  2.24s/it, loss=0.2567, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 623/2855 [26:37<1:22:33,  2.22s/it, loss=0.2567, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 623/2855 [26:39<1:22:33,  2.22s/it, loss=0.1637, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 624/2855 [26:39<1:22:36,  2.22s/it, loss=0.1637, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 624/2855 [26:42<1:22:36,  2.22s/it, loss=0.0790, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 625/2855 [26:42<1:28:05,  2.37s/it, loss=0.0790, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 625/2855 [26:48<1:28:05,  2.37s/it, loss=0.3853, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 626/2855 [26:48<2:12:37,  3.57s/it, loss=0.3853, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 626/2855 [26:51<2:12:37,  3.57s/it, loss=0.1369, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 627/2855 [26:51<2:06:48,  3.41s/it, loss=0.1369, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 627/2855 [26:53<2:06:48,  3.41s/it, loss=0.1016, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 628/2855 [26:53<1:49:50,  2.96s/it, loss=0.1016, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 628/2855 [26:55<1:49:50,  2.96s/it, loss=0.1860, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 629/2855 [26:55<1:42:55,  2.77s/it, loss=0.1860, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 629/2855 [26:58<1:42:55,  2.77s/it, loss=0.2244, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 630/2855 [26:58<1:42:15,  2.76s/it, loss=0.2244, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 630/2855 [27:00<1:42:15,  2.76s/it, loss=0.1992, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 631/2855 [27:00<1:34:51,  2.56s/it, loss=0.1992, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 631/2855 [27:02<1:34:51,  2.56s/it, loss=0.2042, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 632/2855 [27:02<1:27:51,  2.37s/it, loss=0.2042, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 632/2855 [27:04<1:27:51,  2.37s/it, loss=0.1163, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 633/2855 [27:04<1:22:05,  2.22s/it, loss=0.1163, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 633/2855 [27:06<1:22:05,  2.22s/it, loss=0.1480, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 634/2855 [27:06<1:18:11,  2.11s/it, loss=0.1480, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 634/2855 [27:08<1:18:11,  2.11s/it, loss=0.2902, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 635/2855 [27:08<1:23:20,  2.25s/it, loss=0.2902, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 635/2855 [27:55<1:23:20,  2.25s/it, loss=0.1118, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 636/2855 [27:55<9:35:06, 15.55s/it, loss=0.1118, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 636/2855 [27:59<9:35:06, 15.55s/it, loss=0.2172, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 637/2855 [27:59<7:23:33, 12.00s/it, loss=0.2172, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 637/2855 [28:01<7:23:33, 12.00s/it, loss=0.1670, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 638/2855 [28:01<5:31:44,  8.98s/it, loss=0.1670, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 638/2855 [28:03<5:31:44,  8.98s/it, loss=0.1775, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 639/2855 [28:03<4:18:43,  7.01s/it, loss=0.1775, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 639/2855 [28:06<4:18:43,  7.01s/it, loss=0.0567, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 640/2855 [28:06<3:28:57,  5.66s/it, loss=0.0567, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 640/2855 [28:08<3:28:57,  5.66s/it, loss=0.1612, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 641/2855 [28:08<2:54:17,  4.72s/it, loss=0.1612, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 641/2855 [28:11<2:54:17,  4.72s/it, loss=0.2736, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 642/2855 [28:11<2:30:11,  4.07s/it, loss=0.2736, lr=2.99e-05]
Epoch 1/8:  22%|██▏       | 642/2855 [28:13<2:30:11,  4.07s/it, loss=0.2396, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 643/2855 [28:13<2:12:45,  3.60s/it, loss=0.2396, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 643/2855 [28:16<2:12:45,  3.60s/it, loss=0.1767, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 644/2855 [28:16<2:00:21,  3.27s/it, loss=0.1767, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 644/2855 [28:18<2:00:21,  3.27s/it, loss=0.1608, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 645/2855 [28:18<1:49:58,  2.99s/it, loss=0.1608, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 645/2855 [28:20<1:49:58,  2.99s/it, loss=0.1224, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 646/2855 [28:20<1:42:58,  2.80s/it, loss=0.1224, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 646/2855 [28:23<1:42:58,  2.80s/it, loss=0.1475, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 647/2855 [28:23<1:37:20,  2.65s/it, loss=0.1475, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 647/2855 [28:25<1:37:20,  2.65s/it, loss=0.1590, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 648/2855 [28:25<1:34:18,  2.56s/it, loss=0.1590, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 648/2855 [28:27<1:34:18,  2.56s/it, loss=0.1546, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 649/2855 [28:27<1:31:42,  2.49s/it, loss=0.1546, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 649/2855 [28:30<1:31:42,  2.49s/it, loss=0.4484, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 650/2855 [28:30<1:31:33,  2.49s/it, loss=0.4484, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 650/2855 [28:32<1:31:33,  2.49s/it, loss=0.2308, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 651/2855 [28:32<1:29:46,  2.44s/it, loss=0.2308, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 651/2855 [28:35<1:29:46,  2.44s/it, loss=0.0616, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 652/2855 [28:35<1:28:47,  2.42s/it, loss=0.0616, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 652/2855 [28:37<1:28:47,  2.42s/it, loss=0.1593, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 653/2855 [28:37<1:28:33,  2.41s/it, loss=0.1593, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 653/2855 [28:39<1:28:33,  2.41s/it, loss=0.1031, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 654/2855 [28:39<1:27:48,  2.39s/it, loss=0.1031, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 654/2855 [28:43<1:27:48,  2.39s/it, loss=0.2826, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 655/2855 [28:43<1:36:43,  2.64s/it, loss=0.2826, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 655/2855 [28:45<1:36:43,  2.64s/it, loss=0.1324, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 656/2855 [28:45<1:34:18,  2.57s/it, loss=0.1324, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 656/2855 [28:48<1:34:18,  2.57s/it, loss=0.2079, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 657/2855 [28:48<1:34:46,  2.59s/it, loss=0.2079, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 657/2855 [28:50<1:34:46,  2.59s/it, loss=0.2059, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 658/2855 [28:50<1:31:52,  2.51s/it, loss=0.2059, lr=2.99e-05]
Epoch 1/8:  23%|█���▎       | 658/2855 [28:52<1:31:52,  2.51s/it, loss=0.0812, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 659/2855 [28:52<1:32:53,  2.54s/it, loss=0.0812, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 659/2855 [28:55<1:32:53,  2.54s/it, loss=0.2363, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 660/2855 [28:55<1:31:57,  2.51s/it, loss=0.2363, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 660/2855 [28:57<1:31:57,  2.51s/it, loss=0.2582, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 661/2855 [28:57<1:29:58,  2.46s/it, loss=0.2582, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 661/2855 [29:00<1:29:58,  2.46s/it, loss=0.1994, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 662/2855 [29:00<1:30:11,  2.47s/it, loss=0.1994, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 662/2855 [29:02<1:30:11,  2.47s/it, loss=0.3761, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 663/2855 [29:02<1:29:51,  2.46s/it, loss=0.3761, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 663/2855 [29:05<1:29:51,  2.46s/it, loss=0.1838, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 664/2855 [29:05<1:30:20,  2.47s/it, loss=0.1838, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 664/2855 [29:07<1:30:20,  2.47s/it, loss=0.2337, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 665/2855 [29:07<1:29:37,  2.46s/it, loss=0.2337, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 665/2855 [29:10<1:29:37,  2.46s/it, loss=0.1987, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 666/2855 [29:10<1:29:21,  2.45s/it, loss=0.1987, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 666/2855 [29:12<1:29:21,  2.45s/it, loss=0.2787, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 667/2855 [29:12<1:28:17,  2.42s/it, loss=0.2787, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 667/2855 [29:14<1:28:17,  2.42s/it, loss=0.3755, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 668/2855 [29:14<1:27:18,  2.40s/it, loss=0.3755, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 668/2855 [29:17<1:27:18,  2.40s/it, loss=0.2104, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 669/2855 [29:17<1:26:29,  2.37s/it, loss=0.2104, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 669/2855 [29:19<1:26:29,  2.37s/it, loss=0.1629, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 670/2855 [29:19<1:25:29,  2.35s/it, loss=0.1629, lr=2.99e-05]
Epoch 1/8:  23%|██▎       | 670/2855 [29:21<1:25:29,  2.35s/it, loss=0.2690, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 671/2855 [29:21<1:25:04,  2.34s/it, loss=0.2690, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 671/2855 [29:24<1:25:04,  2.34s/it, loss=0.1199, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 672/2855 [29:24<1:26:14,  2.37s/it, loss=0.1199, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 672/2855 [29:26<1:26:14,  2.37s/it, loss=0.1745, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 673/2855 [29:26<1:25:53,  2.36s/it, loss=0.1745, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 673/2855 [29:28<1:25:53,  2.36s/it, loss=0.2135, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 674/2855 [29:28<1:26:31,  2.38s/it, loss=0.2135, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 674/2855 [29:31<1:26:31,  2.38s/it, loss=0.3201, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 675/2855 [29:31<1:25:22,  2.35s/it, loss=0.3201, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 675/2855 [29:37<1:25:22,  2.35s/it, loss=0.1953, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 676/2855 [29:37<2:07:40,  3.52s/it, loss=0.1953, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 676/2855 [29:43<2:07:40,  3.52s/it, loss=0.1988, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 677/2855 [29:43<2:35:59,  4.30s/it, loss=0.1988, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 677/2855 [29:46<2:35:59,  4.30s/it, loss=0.2123, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 678/2855 [29:46<2:19:20,  3.84s/it, loss=0.2123, lr=2.99e-05]
Epoch 1/8:  24%|██▎       | 678/2855 [29:48<2:19:20,  3.84s/it, loss=0.1204, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 679/2855 [29:48<2:02:32,  3.38s/it, loss=0.1204, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 679/2855 [29:50<2:02:32,  3.38s/it, loss=0.1677, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 680/2855 [29:50<1:49:57,  3.03s/it, loss=0.1677, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 680/2855 [29:53<1:49:57,  3.03s/it, loss=0.0558, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 681/2855 [29:53<1:42:18,  2.82s/it, loss=0.0558, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 681/2855 [29:55<1:42:18,  2.82s/it, loss=0.2961, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 682/2855 [29:55<1:36:52,  2.67s/it, loss=0.2961, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 682/2855 [29:58<1:36:52,  2.67s/it, loss=0.2300, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 683/2855 [29:58<1:40:07,  2.77s/it, loss=0.2300, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 683/2855 [30:01<1:40:07,  2.77s/it, loss=0.2330, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 684/2855 [30:01<1:39:55,  2.76s/it, loss=0.2330, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 684/2855 [30:03<1:39:55,  2.76s/it, loss=0.0568, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 685/2855 [30:03<1:39:58,  2.76s/it, loss=0.0568, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 685/2855 [30:06<1:39:58,  2.76s/it, loss=0.1102, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 686/2855 [30:06<1:39:20,  2.75s/it, loss=0.1102, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 686/2855 [30:09<1:39:20,  2.75s/it, loss=0.2031, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 687/2855 [30:09<1:35:09,  2.63s/it, loss=0.2031, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 687/2855 [30:13<1:35:09,  2.63s/it, loss=0.2110, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 688/2855 [30:13<1:55:15,  3.19s/it, loss=0.2110, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 688/2855 [30:16<1:55:15,  3.19s/it, loss=0.3033, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 689/2855 [30:16<1:50:21,  3.06s/it, loss=0.3033, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 689/2855 [30:18<1:50:21,  3.06s/it, loss=0.1541, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 690/2855 [30:18<1:40:41,  2.79s/it, loss=0.1541, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 690/2855 [30:20<1:40:41,  2.79s/it, loss=0.3846, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 691/2855 [30:20<1:34:46,  2.63s/it, loss=0.3846, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 691/2855 [30:23<1:34:46,  2.63s/it, loss=0.1380, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 692/2855 [30:23<1:32:07,  2.56s/it, loss=0.1380, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 692/2855 [30:25<1:32:07,  2.56s/it, loss=0.2714, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 693/2855 [30:25<1:29:24,  2.48s/it, loss=0.2714, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 693/2855 [30:27<1:29:24,  2.48s/it, loss=0.1777, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 694/2855 [30:27<1:28:53,  2.47s/it, loss=0.1777, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 694/2855 [30:30<1:28:53,  2.47s/it, loss=0.2139, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 695/2855 [30:30<1:30:45,  2.52s/it, loss=0.2139, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 695/2855 [30:32<1:30:45,  2.52s/it, loss=0.2611, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 696/2855 [30:32<1:29:52,  2.50s/it, loss=0.2611, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 696/2855 [30:35<1:29:52,  2.50s/it, loss=0.1928, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 697/2855 [30:35<1:29:04,  2.48s/it, loss=0.1928, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 697/2855 [30:37<1:29:04,  2.48s/it, loss=0.2224, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 698/2855 [30:37<1:27:12,  2.43s/it, loss=0.2224, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 698/2855 [30:40<1:27:12,  2.43s/it, loss=0.2235, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 699/2855 [30:40<1:27:42,  2.44s/it, loss=0.2235, lr=2.99e-05]
Epoch 1/8:  24%|██▍       | 699/2855 [30:42<1:27:42,  2.44s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 700/2855 [30:42<1:26:31,  2.41s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 700/2855 [30:44<1:26:31,  2.41s/it, loss=0.0695, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 701/2855 [30:44<1:26:38,  2.41s/it, loss=0.0695, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 701/2855 [30:47<1:26:38,  2.41s/it, loss=0.1794, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 702/2855 [30:47<1:25:47,  2.39s/it, loss=0.1794, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 702/2855 [30:49<1:25:47,  2.39s/it, loss=0.2566, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 703/2855 [30:49<1:25:05,  2.37s/it, loss=0.2566, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 703/2855 [30:51<1:25:05,  2.37s/it, loss=0.2173, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 704/2855 [30:51<1:24:36,  2.36s/it, loss=0.2173, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 704/2855 [30:54<1:24:36,  2.36s/it, loss=0.2090, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 705/2855 [30:54<1:23:54,  2.34s/it, loss=0.2090, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 705/2855 [30:56<1:23:54,  2.34s/it, loss=0.0051, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 706/2855 [30:56<1:26:09,  2.41s/it, loss=0.0051, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 706/2855 [30:59<1:26:09,  2.41s/it, loss=0.1050, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 707/2855 [30:59<1:26:45,  2.42s/it, loss=0.1050, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 707/2855 [31:01<1:26:45,  2.42s/it, loss=0.2085, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 708/2855 [31:01<1:26:15,  2.41s/it, loss=0.2085, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 708/2855 [31:03<1:26:15,  2.41s/it, loss=0.0009, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 709/2855 [31:04<1:26:00,  2.40s/it, loss=0.0009, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 709/2855 [31:07<1:26:00,  2.40s/it, loss=0.1949, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 710/2855 [31:07<1:35:28,  2.67s/it, loss=0.1949, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 710/2855 [31:12<1:35:28,  2.67s/it, loss=0.0407, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 711/2855 [31:12<2:05:19,  3.51s/it, loss=0.0407, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 711/2855 [31:16<2:05:19,  3.51s/it, loss=0.3186, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 712/2855 [31:16<2:02:31,  3.43s/it, loss=0.3186, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 712/2855 [31:18<2:02:31,  3.43s/it, loss=0.2756, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 713/2855 [31:18<1:53:02,  3.17s/it, loss=0.2756, lr=2.99e-05]
Epoch 1/8:  25%|██▍       | 713/2855 [31:21<1:53:02,  3.17s/it, loss=0.1220, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 714/2855 [31:21<1:46:54,  3.00s/it, loss=0.1220, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 714/2855 [31:28<1:46:54,  3.00s/it, loss=0.3524, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 715/2855 [31:28<2:38:30,  4.44s/it, loss=0.3524, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 715/2855 [31:31<2:38:30,  4.44s/it, loss=0.2211, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 716/2855 [31:31<2:23:10,  4.02s/it, loss=0.2211, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 716/2855 [31:34<2:23:10,  4.02s/it, loss=0.3086, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 717/2855 [31:34<2:03:59,  3.48s/it, loss=0.3086, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 717/2855 [31:36<2:03:59,  3.48s/it, loss=0.1058, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 718/2855 [31:36<1:54:31,  3.22s/it, loss=0.1058, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 718/2855 [31:39<1:54:31,  3.22s/it, loss=0.2511, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 719/2855 [31:39<1:46:45,  3.00s/it, loss=0.2511, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 719/2855 [31:41<1:46:45,  3.00s/it, loss=0.1105, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 720/2855 [31:41<1:39:36,  2.80s/it, loss=0.1105, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 720/2855 [31:44<1:39:36,  2.80s/it, loss=0.1083, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 721/2855 [31:44<1:34:50,  2.67s/it, loss=0.1083, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 721/2855 [31:46<1:34:50,  2.67s/it, loss=0.2182, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 722/2855 [31:46<1:32:00,  2.59s/it, loss=0.2182, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 722/2855 [31:48<1:32:00,  2.59s/it, loss=0.1147, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 723/2855 [31:48<1:29:58,  2.53s/it, loss=0.1147, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 723/2855 [31:51<1:29:58,  2.53s/it, loss=0.2299, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 724/2855 [31:51<1:28:30,  2.49s/it, loss=0.2299, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 724/2855 [31:53<1:28:30,  2.49s/it, loss=0.3267, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 725/2855 [31:53<1:28:02,  2.48s/it, loss=0.3267, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 725/2855 [31:56<1:28:02,  2.48s/it, loss=0.2524, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 726/2855 [31:56<1:28:27,  2.49s/it, loss=0.2524, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 726/2855 [31:58<1:28:27,  2.49s/it, loss=0.2885, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 727/2855 [31:58<1:26:59,  2.45s/it, loss=0.2885, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 727/2855 [32:00<1:26:59,  2.45s/it, loss=0.2092, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 728/2855 [32:00<1:24:49,  2.39s/it, loss=0.2092, lr=2.99e-05]
Epoch 1/8:  25%|██▌       | 728/2855 [32:03<1:24:49,  2.39s/it, loss=0.1320, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 729/2855 [32:03<1:23:26,  2.35s/it, loss=0.1320, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 729/2855 [32:05<1:23:26,  2.35s/it, loss=0.1433, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 730/2855 [32:05<1:23:13,  2.35s/it, loss=0.1433, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 730/2855 [32:07<1:23:13,  2.35s/it, loss=0.1712, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 731/2855 [32:07<1:21:36,  2.31s/it, loss=0.1712, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 731/2855 [32:09<1:21:36,  2.31s/it, loss=0.2582, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 732/2855 [32:09<1:21:17,  2.30s/it, loss=0.2582, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 732/2855 [32:12<1:21:17,  2.30s/it, loss=0.3493, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 733/2855 [32:12<1:21:59,  2.32s/it, loss=0.3493, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 733/2855 [32:14<1:21:59,  2.32s/it, loss=0.0962, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 734/2855 [32:14<1:20:15,  2.27s/it, loss=0.0962, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 734/2855 [32:16<1:20:15,  2.27s/it, loss=0.2276, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 735/2855 [32:16<1:22:01,  2.32s/it, loss=0.2276, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 735/2855 [32:19<1:22:01,  2.32s/it, loss=0.1166, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 736/2855 [32:19<1:24:05,  2.38s/it, loss=0.1166, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 736/2855 [32:22<1:24:05,  2.38s/it, loss=0.0746, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 737/2855 [32:22<1:29:25,  2.53s/it, loss=0.0746, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 737/2855 [32:24<1:29:25,  2.53s/it, loss=0.4725, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 738/2855 [32:24<1:28:02,  2.50s/it, loss=0.4725, lr=2.99e-05]
Epoch 1/8:  26%|█���▌       | 738/2855 [32:30<1:28:02,  2.50s/it, loss=0.2539, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 739/2855 [32:30<2:07:40,  3.62s/it, loss=0.2539, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 739/2855 [32:36<2:07:40,  3.62s/it, loss=0.1654, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 740/2855 [32:36<2:31:37,  4.30s/it, loss=0.1654, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 740/2855 [32:39<2:31:37,  4.30s/it, loss=0.1300, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 741/2855 [32:39<2:18:48,  3.94s/it, loss=0.1300, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 741/2855 [32:42<2:18:48,  3.94s/it, loss=0.0341, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 742/2855 [32:42<2:00:24,  3.42s/it, loss=0.0341, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 742/2855 [32:44<2:00:24,  3.42s/it, loss=0.2594, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 743/2855 [32:44<1:51:54,  3.18s/it, loss=0.2594, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 743/2855 [32:47<1:51:54,  3.18s/it, loss=0.3586, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 744/2855 [32:47<1:44:10,  2.96s/it, loss=0.3586, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 744/2855 [32:49<1:44:10,  2.96s/it, loss=0.3449, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 745/2855 [32:49<1:39:22,  2.83s/it, loss=0.3449, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 745/2855 [32:52<1:39:22,  2.83s/it, loss=0.3257, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 746/2855 [32:52<1:36:28,  2.74s/it, loss=0.3257, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 746/2855 [32:54<1:36:28,  2.74s/it, loss=0.2166, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 747/2855 [32:54<1:32:25,  2.63s/it, loss=0.2166, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 747/2855 [32:56<1:32:25,  2.63s/it, loss=0.1619, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 748/2855 [32:56<1:29:37,  2.55s/it, loss=0.1619, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 748/2855 [33:00<1:29:37,  2.55s/it, loss=0.1991, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 749/2855 [33:00<1:35:40,  2.73s/it, loss=0.1991, lr=2.99e-05]
Epoch 1/8:  26%|██▌       | 749/2855 [33:02<1:35:40,  2.73s/it, loss=0.2393, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 750/2855 [33:02<1:31:38,  2.61s/it, loss=0.2393, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 750/2855 [33:05<1:31:38,  2.61s/it, loss=0.1637, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 751/2855 [33:05<1:32:13,  2.63s/it, loss=0.1637, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 751/2855 [33:07<1:32:13,  2.63s/it, loss=0.2079, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 752/2855 [33:07<1:28:56,  2.54s/it, loss=0.2079, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 752/2855 [33:09<1:28:56,  2.54s/it, loss=0.1738, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 753/2855 [33:09<1:26:05,  2.46s/it, loss=0.1738, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 753/2855 [33:12<1:26:05,  2.46s/it, loss=0.1641, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 754/2855 [33:12<1:25:40,  2.45s/it, loss=0.1641, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 754/2855 [33:17<1:25:40,  2.45s/it, loss=0.2045, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 755/2855 [33:17<2:01:07,  3.46s/it, loss=0.2045, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 755/2855 [33:19<2:01:07,  3.46s/it, loss=0.1470, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 756/2855 [33:19<1:45:50,  3.03s/it, loss=0.1470, lr=2.99e-05]
Epoch 1/8:  26%|██▋       | 756/2855 [33:22<1:45:50,  3.03s/it, loss=0.2783, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 757/2855 [33:22<1:40:03,  2.86s/it, loss=0.2783, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 757/2855 [33:24<1:40:03,  2.86s/it, loss=0.2472, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 758/2855 [33:24<1:35:49,  2.74s/it, loss=0.2472, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 758/2855 [33:29<1:35:49,  2.74s/it, loss=0.1582, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 759/2855 [33:29<1:55:11,  3.30s/it, loss=0.1582, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 759/2855 [33:32<1:55:11,  3.30s/it, loss=0.3532, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 760/2855 [33:32<1:52:30,  3.22s/it, loss=0.3532, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 760/2855 [33:35<1:52:30,  3.22s/it, loss=0.3841, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 761/2855 [33:35<1:48:13,  3.10s/it, loss=0.3841, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 761/2855 [33:37<1:48:13,  3.10s/it, loss=0.2398, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 762/2855 [33:37<1:43:01,  2.95s/it, loss=0.2398, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 762/2855 [33:40<1:43:01,  2.95s/it, loss=0.1681, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 763/2855 [33:40<1:39:19,  2.85s/it, loss=0.1681, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 763/2855 [33:43<1:39:19,  2.85s/it, loss=0.1411, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 764/2855 [33:43<1:44:17,  2.99s/it, loss=0.1411, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 764/2855 [33:50<1:44:17,  2.99s/it, loss=0.1615, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 765/2855 [33:50<2:20:55,  4.05s/it, loss=0.1615, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 765/2855 [33:54<2:20:55,  4.05s/it, loss=0.2562, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 766/2855 [33:54<2:20:05,  4.02s/it, loss=0.2562, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 766/2855 [33:57<2:20:05,  4.02s/it, loss=0.6101, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 767/2855 [33:57<2:08:43,  3.70s/it, loss=0.6101, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 767/2855 [34:00<2:08:43,  3.70s/it, loss=0.1856, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 768/2855 [34:00<1:58:54,  3.42s/it, loss=0.1856, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 768/2855 [34:02<1:58:54,  3.42s/it, loss=0.2088, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 769/2855 [34:02<1:52:39,  3.24s/it, loss=0.2088, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 769/2855 [34:05<1:52:39,  3.24s/it, loss=0.1789, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 770/2855 [34:05<1:46:21,  3.06s/it, loss=0.1789, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 770/2855 [34:08<1:46:21,  3.06s/it, loss=0.1970, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 771/2855 [34:08<1:45:27,  3.04s/it, loss=0.1970, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 771/2855 [34:11<1:45:27,  3.04s/it, loss=0.2149, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 772/2855 [34:11<1:39:53,  2.88s/it, loss=0.2149, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 772/2855 [34:14<1:39:53,  2.88s/it, loss=0.1579, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 773/2855 [34:14<1:41:23,  2.92s/it, loss=0.1579, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 773/2855 [36:18<1:41:23,  2.92s/it, loss=0.2272, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 774/2855 [36:18<22:41:10, 39.25s/it, loss=0.2272, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 774/2855 [36:53<22:41:10, 39.25s/it, loss=0.1037, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 775/2855 [36:53<21:56:38, 37.98s/it, loss=0.1037, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 775/2855 [36:54<21:56:38, 37.98s/it, loss=0.1889, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 776/2855 [36:54<15:36:20, 27.02s/it, loss=0.1889, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 776/2855 [36:56<15:36:20, 27.02s/it, loss=0.1597, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 777/2855 [36:56<11:10:15, 19.35s/it, loss=0.1597, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 777/2855 [37:02<11:10:15, 19.35s/it, loss=0.2896, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 778/2855 [37:02<8:51:39, 15.36s/it, loss=0.2896, lr=2.99e-05] 
Epoch 1/8:  27%|██▋       | 778/2855 [37:04<8:51:39, 15.36s/it, loss=0.0479, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 779/2855 [37:04<6:40:06, 11.56s/it, loss=0.0479, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 779/2855 [37:07<6:40:06, 11.56s/it, loss=0.2213, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 780/2855 [37:07<5:04:59,  8.82s/it, loss=0.2213, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 780/2855 [37:09<5:04:59,  8.82s/it, loss=0.1402, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 781/2855 [37:09<4:00:27,  6.96s/it, loss=0.1402, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 781/2855 [37:12<4:00:27,  6.96s/it, loss=0.1290, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 782/2855 [37:12<3:13:09,  5.59s/it, loss=0.1290, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 782/2855 [37:14<3:13:09,  5.59s/it, loss=0.2545, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 783/2855 [37:14<2:42:20,  4.70s/it, loss=0.2545, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 783/2855 [37:16<2:42:20,  4.70s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 784/2855 [37:16<2:16:09,  3.94s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 784/2855 [37:20<2:16:09,  3.94s/it, loss=0.0992, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 785/2855 [37:20<2:07:12,  3.69s/it, loss=0.0992, lr=2.99e-05]
Epoch 1/8:  27%|██▋       | 785/2855 [37:22<2:07:12,  3.69s/it, loss=0.2662, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 786/2855 [37:22<1:54:57,  3.33s/it, loss=0.2662, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 786/2855 [37:25<1:54:57,  3.33s/it, loss=0.2025, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 787/2855 [37:25<1:49:26,  3.18s/it, loss=0.2025, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 787/2855 [37:28<1:49:26,  3.18s/it, loss=0.2630, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 788/2855 [37:28<1:43:46,  3.01s/it, loss=0.2630, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 788/2855 [37:30<1:43:46,  3.01s/it, loss=0.3054, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 789/2855 [37:30<1:40:31,  2.92s/it, loss=0.3054, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 789/2855 [37:33<1:40:31,  2.92s/it, loss=0.2865, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 790/2855 [37:33<1:38:50,  2.87s/it, loss=0.2865, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 790/2855 [37:36<1:38:50,  2.87s/it, loss=0.2866, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 791/2855 [37:36<1:36:08,  2.79s/it, loss=0.2866, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 791/2855 [37:38<1:36:08,  2.79s/it, loss=0.1339, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 792/2855 [37:38<1:33:11,  2.71s/it, loss=0.1339, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 792/2855 [37:41<1:33:11,  2.71s/it, loss=0.1464, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 793/2855 [37:41<1:32:12,  2.68s/it, loss=0.1464, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 793/2855 [38:12<1:32:12,  2.68s/it, loss=0.1581, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 794/2855 [38:12<6:23:21, 11.16s/it, loss=0.1581, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 794/2855 [38:13<6:23:21, 11.16s/it, loss=0.2313, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 795/2855 [38:13<4:46:46,  8.35s/it, loss=0.2313, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 795/2855 [38:15<4:46:46,  8.35s/it, loss=0.1719, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 796/2855 [38:15<3:36:19,  6.30s/it, loss=0.1719, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 796/2855 [38:19<3:36:19,  6.30s/it, loss=0.1191, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 797/2855 [38:19<3:15:07,  5.69s/it, loss=0.1191, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 797/2855 [38:22<3:15:07,  5.69s/it, loss=0.1567, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 798/2855 [38:22<2:40:54,  4.69s/it, loss=0.1567, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 798/2855 [38:24<2:40:54,  4.69s/it, loss=0.1095, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 799/2855 [38:24<2:16:51,  3.99s/it, loss=0.1095, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 799/2855 [38:26<2:16:51,  3.99s/it, loss=0.2338, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 800/2855 [38:26<2:00:22,  3.51s/it, loss=0.2338, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 800/2855 [38:29<2:00:22,  3.51s/it, loss=0.1509, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 801/2855 [38:29<1:55:29,  3.37s/it, loss=0.1509, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 801/2855 [38:32<1:55:29,  3.37s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 802/2855 [38:32<1:46:58,  3.13s/it, loss=0.1788, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 802/2855 [38:35<1:46:58,  3.13s/it, loss=0.0863, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 803/2855 [38:35<1:43:27,  3.03s/it, loss=0.0863, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 803/2855 [38:37<1:43:27,  3.03s/it, loss=0.1453, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 804/2855 [38:37<1:38:18,  2.88s/it, loss=0.1453, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 804/2855 [38:40<1:38:18,  2.88s/it, loss=0.3912, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 805/2855 [38:40<1:35:26,  2.79s/it, loss=0.3912, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 805/2855 [38:43<1:35:26,  2.79s/it, loss=0.2392, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 806/2855 [38:43<1:33:51,  2.75s/it, loss=0.2392, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 806/2855 [38:50<1:33:51,  2.75s/it, loss=0.0550, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 807/2855 [38:50<2:18:11,  4.05s/it, loss=0.0550, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 807/2855 [38:56<2:18:11,  4.05s/it, loss=0.3380, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 808/2855 [38:56<2:37:05,  4.60s/it, loss=0.3380, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 808/2855 [39:00<2:37:05,  4.60s/it, loss=0.1259, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 809/2855 [39:00<2:31:12,  4.43s/it, loss=0.1259, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 809/2855 [39:04<2:31:12,  4.43s/it, loss=0.3453, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 810/2855 [39:04<2:27:57,  4.34s/it, loss=0.3453, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 810/2855 [39:08<2:27:57,  4.34s/it, loss=0.2096, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 811/2855 [39:08<2:26:04,  4.29s/it, loss=0.2096, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 811/2855 [39:12<2:26:04,  4.29s/it, loss=0.2451, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 812/2855 [39:12<2:23:55,  4.23s/it, loss=0.2451, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 812/2855 [39:15<2:23:55,  4.23s/it, loss=0.2490, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 813/2855 [39:15<2:10:01,  3.82s/it, loss=0.2490, lr=2.99e-05]
Epoch 1/8:  28%|██▊       | 813/2855 [39:17<2:10:01,  3.82s/it, loss=0.0945, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 814/2855 [39:17<1:58:21,  3.48s/it, loss=0.0945, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 814/2855 [39:20<1:58:21,  3.48s/it, loss=0.2223, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 815/2855 [39:20<1:49:28,  3.22s/it, loss=0.2223, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 815/2855 [39:23<1:49:28,  3.22s/it, loss=0.1201, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 816/2855 [39:23<1:45:26,  3.10s/it, loss=0.1201, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 816/2855 [39:26<1:45:26,  3.10s/it, loss=0.1638, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 817/2855 [39:26<1:42:49,  3.03s/it, loss=0.1638, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 817/2855 [39:29<1:42:49,  3.03s/it, loss=0.2264, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 818/2855 [39:29<1:39:44,  2.94s/it, loss=0.2264, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 818/2855 [39:31<1:39:44,  2.94s/it, loss=0.3412, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 819/2855 [39:31<1:36:50,  2.85s/it, loss=0.3412, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 819/2855 [39:34<1:36:50,  2.85s/it, loss=0.2364, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 820/2855 [39:34<1:32:49,  2.74s/it, loss=0.2364, lr=2.99e-05]
Epoch 1/8:  29%|██▊       | 820/2855 [39:36<1:32:49,  2.74s/it, loss=0.0686, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 821/2855 [39:36<1:30:54,  2.68s/it, loss=0.0686, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 821/2855 [39:39<1:30:54,  2.68s/it, loss=0.2200, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 822/2855 [39:39<1:29:14,  2.63s/it, loss=0.2200, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 822/2855 [39:41<1:29:14,  2.63s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 823/2855 [39:41<1:26:35,  2.56s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 823/2855 [39:44<1:26:35,  2.56s/it, loss=0.0960, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 824/2855 [39:44<1:27:53,  2.60s/it, loss=0.0960, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 824/2855 [39:46<1:27:53,  2.60s/it, loss=0.1763, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 825/2855 [39:46<1:26:05,  2.54s/it, loss=0.1763, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 825/2855 [39:49<1:26:05,  2.54s/it, loss=0.1949, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 826/2855 [39:49<1:33:43,  2.77s/it, loss=0.1949, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 826/2855 [39:53<1:33:43,  2.77s/it, loss=0.1381, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 827/2855 [39:53<1:44:25,  3.09s/it, loss=0.1381, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 827/2855 [39:59<1:44:25,  3.09s/it, loss=0.0911, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 828/2855 [39:59<2:14:01,  3.97s/it, loss=0.0911, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 828/2855 [40:03<2:14:01,  3.97s/it, loss=0.1278, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 829/2855 [40:03<2:07:39,  3.78s/it, loss=0.1278, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 829/2855 [40:05<2:07:39,  3.78s/it, loss=0.1635, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 830/2855 [40:05<1:55:20,  3.42s/it, loss=0.1635, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 830/2855 [40:08<1:55:20,  3.42s/it, loss=0.1687, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 831/2855 [40:08<1:48:03,  3.20s/it, loss=0.1687, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 831/2855 [40:11<1:48:03,  3.20s/it, loss=0.2609, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 832/2855 [40:11<1:42:15,  3.03s/it, loss=0.2609, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 832/2855 [40:13<1:42:15,  3.03s/it, loss=0.0468, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 833/2855 [40:13<1:40:27,  2.98s/it, loss=0.0468, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 833/2855 [41:07<1:40:27,  2.98s/it, loss=0.3142, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 834/2855 [41:07<10:11:01, 18.14s/it, loss=0.3142, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 834/2855 [41:08<10:11:01, 18.14s/it, loss=0.1588, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 835/2855 [41:08<7:20:54, 13.10s/it, loss=0.1588, lr=2.99e-05] 
Epoch 1/8:  29%|██▉       | 835/2855 [41:10<7:20:54, 13.10s/it, loss=0.1608, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 836/2855 [41:10<5:23:10,  9.60s/it, loss=0.1608, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 836/2855 [41:14<5:23:10,  9.60s/it, loss=0.1515, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 837/2855 [41:14<4:26:47,  7.93s/it, loss=0.1515, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 837/2855 [41:17<4:26:47,  7.93s/it, loss=0.2321, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 838/2855 [41:17<3:35:36,  6.41s/it, loss=0.2321, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 838/2855 [41:19<3:35:36,  6.41s/it, loss=0.3200, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 839/2855 [41:19<2:55:51,  5.23s/it, loss=0.3200, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 839/2855 [41:21<2:55:51,  5.23s/it, loss=0.2562, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 840/2855 [41:21<2:25:34,  4.33s/it, loss=0.2562, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 840/2855 [41:23<2:25:34,  4.33s/it, loss=0.1633, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 841/2855 [41:23<2:02:51,  3.66s/it, loss=0.1633, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 841/2855 [41:26<2:02:51,  3.66s/it, loss=0.1974, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 842/2855 [41:26<1:56:00,  3.46s/it, loss=0.1974, lr=2.99e-05]
Epoch 1/8:  29%|██▉       | 842/2855 [41:30<1:56:00,  3.46s/it, loss=0.0621, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 843/2855 [41:30<1:59:21,  3.56s/it, loss=0.0621, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 843/2855 [41:33<1:59:21,  3.56s/it, loss=0.2291, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 844/2855 [41:33<1:52:33,  3.36s/it, loss=0.2291, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 844/2855 [41:38<1:52:33,  3.36s/it, loss=0.2255, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 845/2855 [41:38<2:12:11,  3.95s/it, loss=0.2255, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 845/2855 [41:42<2:12:11,  3.95s/it, loss=0.2616, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 846/2855 [41:42<2:04:19,  3.71s/it, loss=0.2616, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 846/2855 [41:44<2:04:19,  3.71s/it, loss=0.2537, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 847/2855 [41:44<1:52:10,  3.35s/it, loss=0.2537, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 847/2855 [41:48<1:52:10,  3.35s/it, loss=0.3653, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 848/2855 [41:48<1:52:35,  3.37s/it, loss=0.3653, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 848/2855 [41:50<1:52:35,  3.37s/it, loss=0.0284, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 849/2855 [41:50<1:47:58,  3.23s/it, loss=0.0284, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 849/2855 [41:54<1:47:58,  3.23s/it, loss=0.2118, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 850/2855 [41:54<1:53:33,  3.40s/it, loss=0.2118, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 850/2855 [41:58<1:53:33,  3.40s/it, loss=0.2868, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 851/2855 [41:58<1:59:23,  3.57s/it, loss=0.2868, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 851/2855 [42:02<1:59:23,  3.57s/it, loss=0.2512, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 852/2855 [42:02<1:58:52,  3.56s/it, loss=0.2512, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 852/2855 [42:04<1:58:52,  3.56s/it, loss=0.2202, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 853/2855 [42:04<1:47:24,  3.22s/it, loss=0.2202, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 853/2855 [42:08<1:47:24,  3.22s/it, loss=0.2119, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 854/2855 [42:08<1:49:01,  3.27s/it, loss=0.2119, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 854/2855 [42:10<1:49:01,  3.27s/it, loss=0.2547, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 855/2855 [42:10<1:39:31,  2.99s/it, loss=0.2547, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 855/2855 [42:13<1:39:31,  2.99s/it, loss=0.0796, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 856/2855 [42:13<1:42:05,  3.06s/it, loss=0.0796, lr=2.99e-05]
Epoch 1/8:  30%|██▉       | 856/2855 [42:16<1:42:05,  3.06s/it, loss=0.2123, lr=2.99e-05]
Epoch 1/8:  30%|███       | 857/2855 [42:16<1:37:00,  2.91s/it, loss=0.2123, lr=2.99e-05]
Epoch 1/8:  30%|███       | 857/2855 [42:19<1:37:00,  2.91s/it, loss=0.0506, lr=2.99e-05]
Epoch 1/8:  30%|███       | 858/2855 [42:19<1:37:36,  2.93s/it, loss=0.0506, lr=2.99e-05]
Epoch 1/8:  30%|███       | 858/2855 [42:21<1:37:36,  2.93s/it, loss=0.1322, lr=2.99e-05]
Epoch 1/8:  30%|███       | 859/2855 [42:21<1:34:26,  2.84s/it, loss=0.1322, lr=2.99e-05]
Epoch 1/8:  30%|███       | 859/2855 [42:24<1:34:26,  2.84s/it, loss=0.0911, lr=2.99e-05]
Epoch 1/8:  30%|███       | 860/2855 [42:24<1:32:26,  2.78s/it, loss=0.0911, lr=2.99e-05]
Epoch 1/8:  30%|███       | 860/2855 [42:26<1:32:26,  2.78s/it, loss=0.2008, lr=2.99e-05]
Epoch 1/8:  30%|███       | 861/2855 [42:26<1:29:28,  2.69s/it, loss=0.2008, lr=2.99e-05]
Epoch 1/8:  30%|███       | 861/2855 [42:29<1:29:28,  2.69s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  30%|███       | 862/2855 [42:29<1:31:44,  2.76s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  30%|███       | 862/2855 [43:40<1:31:44,  2.76s/it, loss=0.2615, lr=2.99e-05]
Epoch 1/8:  30%|███       | 863/2855 [43:40<12:49:14, 23.17s/it, loss=0.2615, lr=2.99e-05]
Epoch 1/8:  30%|███       | 863/2855 [43:42<12:49:14, 23.17s/it, loss=0.2336, lr=2.99e-05]
Epoch 1/8:  30%|███       | 864/2855 [43:42<9:20:34, 16.89s/it, loss=0.2336, lr=2.99e-05] 
Epoch 1/8:  30%|███       | 864/2855 [43:46<9:20:34, 16.89s/it, loss=0.0559, lr=2.99e-05]
Epoch 1/8:  30%|███       | 865/2855 [43:46<7:06:52, 12.87s/it, loss=0.0559, lr=2.99e-05]
Epoch 1/8:  30%|███       | 865/2855 [43:48<7:06:52, 12.87s/it, loss=0.0823, lr=2.99e-05]
Epoch 1/8:  30%|███       | 866/2855 [43:48<5:21:16,  9.69s/it, loss=0.0823, lr=2.99e-05]
Epoch 1/8:  30%|███       | 866/2855 [43:51<5:21:16,  9.69s/it, loss=0.1188, lr=2.99e-05]
Epoch 1/8:  30%|███       | 867/2855 [43:51<4:10:54,  7.57s/it, loss=0.1188, lr=2.99e-05]
Epoch 1/8:  30%|███       | 867/2855 [43:53<4:10:54,  7.57s/it, loss=0.2884, lr=2.99e-05]
Epoch 1/8:  30%|███       | 868/2855 [43:53<3:18:31,  5.99s/it, loss=0.2884, lr=2.99e-05]
Epoch 1/8:  30%|███       | 868/2855 [43:56<3:18:31,  5.99s/it, loss=0.2795, lr=2.99e-05]
Epoch 1/8:  30%|███       | 869/2855 [43:56<2:43:17,  4.93s/it, loss=0.2795, lr=2.99e-05]
Epoch 1/8:  30%|███       | 869/2855 [43:58<2:43:17,  4.93s/it, loss=0.1810, lr=2.99e-05]
Epoch 1/8:  30%|███       | 870/2855 [43:58<2:19:51,  4.23s/it, loss=0.1810, lr=2.99e-05]
Epoch 1/8:  30%|███       | 870/2855 [44:00<2:19:51,  4.23s/it, loss=0.2378, lr=2.99e-05]
Epoch 1/8:  31%|███       | 871/2855 [44:00<2:00:45,  3.65s/it, loss=0.2378, lr=2.99e-05]
Epoch 1/8:  31%|███       | 871/2855 [44:03<2:00:45,  3.65s/it, loss=0.0729, lr=2.99e-05]
Epoch 1/8:  31%|███       | 872/2855 [44:03<1:51:18,  3.37s/it, loss=0.0729, lr=2.99e-05]
Epoch 1/8:  31%|███       | 872/2855 [44:17<1:51:18,  3.37s/it, loss=0.3532, lr=2.99e-05]
Epoch 1/8:  31%|███       | 873/2855 [44:17<3:35:02,  6.51s/it, loss=0.3532, lr=2.99e-05]
Epoch 1/8:  31%|███       | 873/2855 [44:18<3:35:02,  6.51s/it, loss=0.2833, lr=2.99e-05]
Epoch 1/8:  31%|███       | 874/2855 [44:18<2:45:38,  5.02s/it, loss=0.2833, lr=2.99e-05]
Epoch 1/8:  31%|███       | 874/2855 [44:21<2:45:38,  5.02s/it, loss=0.0742, lr=2.99e-05]
Epoch 1/8:  31%|███       | 875/2855 [44:21<2:17:22,  4.16s/it, loss=0.0742, lr=2.99e-05]
Epoch 1/8:  31%|███       | 875/2855 [44:25<2:17:22,  4.16s/it, loss=0.3929, lr=2.99e-05]
Epoch 1/8:  31%|███       | 876/2855 [44:25<2:23:23,  4.35s/it, loss=0.3929, lr=2.99e-05]
Epoch 1/8:  31%|███       | 876/2855 [44:29<2:23:23,  4.35s/it, loss=0.2242, lr=2.99e-05]
Epoch 1/8:  31%|███       | 877/2855 [44:29<2:13:40,  4.05s/it, loss=0.2242, lr=2.99e-05]
Epoch 1/8:  31%|███       | 877/2855 [44:31<2:13:40,  4.05s/it, loss=0.0434, lr=2.99e-05]
Epoch 1/8:  31%|███       | 878/2855 [44:31<1:58:02,  3.58s/it, loss=0.0434, lr=2.99e-05]
Epoch 1/8:  31%|███       | 878/2855 [44:34<1:58:02,  3.58s/it, loss=0.0971, lr=2.99e-05]
Epoch 1/8:  31%|███       | 879/2855 [44:34<1:49:30,  3.33s/it, loss=0.0971, lr=2.99e-05]
Epoch 1/8:  31%|███       | 879/2855 [44:36<1:49:30,  3.33s/it, loss=0.4176, lr=2.99e-05]
Epoch 1/8:  31%|███       | 880/2855 [44:36<1:39:39,  3.03s/it, loss=0.4176, lr=2.99e-05]
Epoch 1/8:  31%|███       | 880/2855 [44:39<1:39:39,  3.03s/it, loss=0.1502, lr=2.99e-05]
Epoch 1/8:  31%|███       | 881/2855 [44:39<1:32:33,  2.81s/it, loss=0.1502, lr=2.99e-05]
Epoch 1/8:  31%|███       | 881/2855 [44:41<1:32:33,  2.81s/it, loss=0.0274, lr=2.99e-05]
Epoch 1/8:  31%|███       | 882/2855 [44:41<1:30:05,  2.74s/it, loss=0.0274, lr=2.99e-05]
Epoch 1/8:  31%|███       | 882/2855 [44:44<1:30:05,  2.74s/it, loss=0.1064, lr=2.99e-05]
Epoch 1/8:  31%|███       | 883/2855 [44:44<1:30:00,  2.74s/it, loss=0.1064, lr=2.99e-05]
Epoch 1/8:  31%|███       | 883/2855 [44:46<1:30:00,  2.74s/it, loss=0.1414, lr=2.99e-05]
Epoch 1/8:  31%|███       | 884/2855 [44:46<1:27:49,  2.67s/it, loss=0.1414, lr=2.99e-05]
Epoch 1/8:  31%|███       | 884/2855 [44:49<1:27:49,  2.67s/it, loss=0.0480, lr=2.99e-05]
Epoch 1/8:  31%|███       | 885/2855 [44:49<1:24:49,  2.58s/it, loss=0.0480, lr=2.99e-05]
Epoch 1/8:  31%|███       | 885/2855 [44:51<1:24:49,  2.58s/it, loss=0.1614, lr=2.99e-05]
Epoch 1/8:  31%|███       | 886/2855 [44:51<1:24:45,  2.58s/it, loss=0.1614, lr=2.99e-05]
Epoch 1/8:  31%|███       | 886/2855 [44:54<1:24:45,  2.58s/it, loss=0.3157, lr=2.99e-05]
Epoch 1/8:  31%|███       | 887/2855 [44:54<1:22:22,  2.51s/it, loss=0.3157, lr=2.99e-05]
Epoch 1/8:  31%|███       | 887/2855 [44:56<1:22:22,  2.51s/it, loss=0.1439, lr=2.99e-05]
Epoch 1/8:  31%|███       | 888/2855 [44:56<1:20:39,  2.46s/it, loss=0.1439, lr=2.99e-05]
Epoch 1/8:  31%|███       | 888/2855 [44:59<1:20:39,  2.46s/it, loss=0.1399, lr=2.99e-05]
Epoch 1/8:  31%|███       | 889/2855 [44:59<1:19:54,  2.44s/it, loss=0.1399, lr=2.99e-05]
Epoch 1/8:  31%|███       | 889/2855 [45:01<1:19:54,  2.44s/it, loss=0.2501, lr=2.99e-05]
Epoch 1/8:  31%|███       | 890/2855 [45:01<1:20:43,  2.47s/it, loss=0.2501, lr=2.99e-05]
Epoch 1/8:  31%|███       | 890/2855 [45:04<1:20:43,  2.47s/it, loss=0.1155, lr=2.99e-05]
Epoch 1/8:  31%|███       | 891/2855 [45:04<1:23:41,  2.56s/it, loss=0.1155, lr=2.99e-05]
Epoch 1/8:  31%|███       | 891/2855 [45:07<1:23:41,  2.56s/it, loss=0.0946, lr=2.99e-05]
Epoch 1/8:  31%|███       | 892/2855 [45:07<1:26:10,  2.63s/it, loss=0.0946, lr=2.99e-05]
Epoch 1/8:  31%|███       | 892/2855 [45:09<1:26:10,  2.63s/it, loss=0.3751, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 893/2855 [45:09<1:26:52,  2.66s/it, loss=0.3751, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 893/2855 [45:12<1:26:52,  2.66s/it, loss=0.2928, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 894/2855 [45:12<1:29:02,  2.72s/it, loss=0.2928, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 894/2855 [45:15<1:29:02,  2.72s/it, loss=0.1570, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 895/2855 [45:15<1:26:26,  2.65s/it, loss=0.1570, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 895/2855 [45:18<1:26:26,  2.65s/it, loss=0.3028, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 896/2855 [45:18<1:30:13,  2.76s/it, loss=0.3028, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 896/2855 [45:21<1:30:13,  2.76s/it, loss=0.1438, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 897/2855 [45:21<1:31:24,  2.80s/it, loss=0.1438, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 897/2855 [45:23<1:31:24,  2.80s/it, loss=0.3248, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 898/2855 [45:23<1:27:38,  2.69s/it, loss=0.3248, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 898/2855 [45:25<1:27:38,  2.69s/it, loss=0.2021, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 899/2855 [45:25<1:24:20,  2.59s/it, loss=0.2021, lr=2.99e-05]
Epoch 1/8:  31%|███▏      | 899/2855 [45:28<1:24:20,  2.59s/it, loss=0.1634, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 900/2855 [45:28<1:28:12,  2.71s/it, loss=0.1634, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 900/2855 [45:31<1:28:12,  2.71s/it, loss=0.2688, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 901/2855 [45:31<1:30:42,  2.79s/it, loss=0.2688, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 901/2855 [45:34<1:30:42,  2.79s/it, loss=0.2785, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 902/2855 [45:34<1:27:54,  2.70s/it, loss=0.2785, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 902/2855 [45:37<1:27:54,  2.70s/it, loss=0.1801, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 903/2855 [45:37<1:28:58,  2.73s/it, loss=0.1801, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 903/2855 [45:39<1:28:58,  2.73s/it, loss=0.2280, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 904/2855 [45:39<1:29:14,  2.74s/it, loss=0.2280, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 904/2855 [45:42<1:29:14,  2.74s/it, loss=0.2005, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 905/2855 [45:42<1:24:28,  2.60s/it, loss=0.2005, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 905/2855 [45:44<1:24:28,  2.60s/it, loss=0.1127, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 906/2855 [45:44<1:26:19,  2.66s/it, loss=0.1127, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 906/2855 [45:47<1:26:19,  2.66s/it, loss=0.3470, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 907/2855 [45:47<1:26:09,  2.65s/it, loss=0.3470, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 907/2855 [45:50<1:26:09,  2.65s/it, loss=0.2479, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 908/2855 [45:50<1:28:41,  2.73s/it, loss=0.2479, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 908/2855 [45:53<1:28:41,  2.73s/it, loss=0.2554, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 909/2855 [45:53<1:30:05,  2.78s/it, loss=0.2554, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 909/2855 [45:55<1:30:05,  2.78s/it, loss=0.3863, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 910/2855 [45:55<1:25:46,  2.65s/it, loss=0.3863, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 910/2855 [45:58<1:25:46,  2.65s/it, loss=0.2006, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 911/2855 [45:58<1:22:40,  2.55s/it, loss=0.2006, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 911/2855 [46:00<1:22:40,  2.55s/it, loss=0.4160, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 912/2855 [46:00<1:21:26,  2.51s/it, loss=0.4160, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 912/2855 [46:09<1:21:26,  2.51s/it, loss=0.2184, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 913/2855 [46:09<2:25:47,  4.50s/it, loss=0.2184, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 913/2855 [46:15<2:25:47,  4.50s/it, loss=0.3247, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 914/2855 [46:15<2:41:31,  4.99s/it, loss=0.3247, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 914/2855 [46:19<2:41:31,  4.99s/it, loss=0.2833, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 915/2855 [46:19<2:27:08,  4.55s/it, loss=0.2833, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 915/2855 [46:21<2:27:08,  4.55s/it, loss=0.1628, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 916/2855 [46:21<2:03:44,  3.83s/it, loss=0.1628, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 916/2855 [46:24<2:03:44,  3.83s/it, loss=0.0800, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 917/2855 [46:24<1:52:05,  3.47s/it, loss=0.0800, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 917/2855 [46:26<1:52:05,  3.47s/it, loss=0.1456, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 918/2855 [46:26<1:41:08,  3.13s/it, loss=0.1456, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 918/2855 [46:28<1:41:08,  3.13s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 919/2855 [46:28<1:34:14,  2.92s/it, loss=0.1489, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 919/2855 [46:31<1:34:14,  2.92s/it, loss=0.2042, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 920/2855 [46:31<1:29:37,  2.78s/it, loss=0.2042, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 920/2855 [46:33<1:29:37,  2.78s/it, loss=0.2099, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 921/2855 [46:33<1:26:28,  2.68s/it, loss=0.2099, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 921/2855 [46:36<1:26:28,  2.68s/it, loss=0.1499, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 922/2855 [46:36<1:24:45,  2.63s/it, loss=0.1499, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 922/2855 [46:38<1:24:45,  2.63s/it, loss=0.1959, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 923/2855 [46:38<1:22:02,  2.55s/it, loss=0.1959, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 923/2855 [46:41<1:22:02,  2.55s/it, loss=0.1411, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 924/2855 [46:41<1:20:57,  2.52s/it, loss=0.1411, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 924/2855 [46:43<1:20:57,  2.52s/it, loss=0.2256, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 925/2855 [46:43<1:22:51,  2.58s/it, loss=0.2256, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 925/2855 [46:46<1:22:51,  2.58s/it, loss=0.1291, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 926/2855 [46:46<1:21:20,  2.53s/it, loss=0.1291, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 926/2855 [46:48<1:21:20,  2.53s/it, loss=0.1777, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 927/2855 [46:48<1:21:22,  2.53s/it, loss=0.1777, lr=2.99e-05]
Epoch 1/8:  32%|███▏      | 927/2855 [46:51<1:21:22,  2.53s/it, loss=0.3771, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 928/2855 [46:51<1:20:57,  2.52s/it, loss=0.3771, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 928/2855 [46:54<1:20:57,  2.52s/it, loss=0.1042, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 929/2855 [46:54<1:25:10,  2.65s/it, loss=0.1042, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 929/2855 [46:56<1:25:10,  2.65s/it, loss=0.3670, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 930/2855 [46:56<1:19:46,  2.49s/it, loss=0.3670, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 930/2855 [46:58<1:19:46,  2.49s/it, loss=0.2593, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 931/2855 [46:58<1:19:59,  2.49s/it, loss=0.2593, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 931/2855 [47:01<1:19:59,  2.49s/it, loss=0.1404, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 932/2855 [47:01<1:22:13,  2.57s/it, loss=0.1404, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 932/2855 [47:03<1:22:13,  2.57s/it, loss=0.3492, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 933/2855 [47:03<1:20:04,  2.50s/it, loss=0.3492, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 933/2855 [47:06<1:20:04,  2.50s/it, loss=0.2140, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 934/2855 [47:06<1:20:25,  2.51s/it, loss=0.2140, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 934/2855 [47:08<1:20:25,  2.51s/it, loss=0.2436, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 935/2855 [47:08<1:18:59,  2.47s/it, loss=0.2436, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 935/2855 [47:11<1:18:59,  2.47s/it, loss=0.2066, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 936/2855 [47:11<1:18:33,  2.46s/it, loss=0.2066, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 936/2855 [47:13<1:18:33,  2.46s/it, loss=0.3466, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 937/2855 [47:13<1:17:15,  2.42s/it, loss=0.3466, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 937/2855 [47:15<1:17:15,  2.42s/it, loss=0.3468, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 938/2855 [47:15<1:16:42,  2.40s/it, loss=0.3468, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 938/2855 [47:18<1:16:42,  2.40s/it, loss=0.1580, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 939/2855 [47:18<1:17:02,  2.41s/it, loss=0.1580, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 939/2855 [47:20<1:17:02,  2.41s/it, loss=0.2346, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 940/2855 [47:20<1:17:46,  2.44s/it, loss=0.2346, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 940/2855 [47:23<1:17:46,  2.44s/it, loss=0.1532, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 941/2855 [47:23<1:17:43,  2.44s/it, loss=0.1532, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 941/2855 [47:25<1:17:43,  2.44s/it, loss=0.2994, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 942/2855 [47:25<1:19:19,  2.49s/it, loss=0.2994, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 942/2855 [47:28<1:19:19,  2.49s/it, loss=0.2731, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 943/2855 [47:28<1:18:23,  2.46s/it, loss=0.2731, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 943/2855 [47:30<1:18:23,  2.46s/it, loss=0.2756, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 944/2855 [47:30<1:16:53,  2.41s/it, loss=0.2756, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 944/2855 [47:33<1:16:53,  2.41s/it, loss=0.3008, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 945/2855 [47:33<1:17:02,  2.42s/it, loss=0.3008, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 945/2855 [47:35<1:17:02,  2.42s/it, loss=0.1527, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 946/2855 [47:35<1:18:24,  2.46s/it, loss=0.1527, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 946/2855 [47:38<1:18:24,  2.46s/it, loss=0.1406, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 947/2855 [47:38<1:19:05,  2.49s/it, loss=0.1406, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 947/2855 [47:40<1:19:05,  2.49s/it, loss=0.2962, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 948/2855 [47:40<1:18:22,  2.47s/it, loss=0.2962, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 948/2855 [47:43<1:18:22,  2.47s/it, loss=0.1294, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 949/2855 [47:43<1:18:22,  2.47s/it, loss=0.1294, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 949/2855 [47:45<1:18:22,  2.47s/it, loss=0.1039, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 950/2855 [47:45<1:18:21,  2.47s/it, loss=0.1039, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 950/2855 [47:48<1:18:21,  2.47s/it, loss=0.0784, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 951/2855 [47:48<1:19:10,  2.50s/it, loss=0.0784, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 951/2855 [47:53<1:19:10,  2.50s/it, loss=0.0186, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 952/2855 [47:53<1:43:24,  3.26s/it, loss=0.0186, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 952/2855 [49:12<1:43:24,  3.26s/it, loss=0.0709, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 953/2855 [49:12<13:44:44, 26.02s/it, loss=0.0709, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 953/2855 [49:13<13:44:44, 26.02s/it, loss=0.1300, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 954/2855 [49:13<9:50:13, 18.63s/it, loss=0.1300, lr=2.99e-05] 
Epoch 1/8:  33%|███▎      | 954/2855 [49:14<9:50:13, 18.63s/it, loss=0.1558, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 955/2855 [49:14<7:05:18, 13.43s/it, loss=0.1558, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 955/2855 [49:16<7:05:18, 13.43s/it, loss=0.1661, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 956/2855 [49:16<5:08:45,  9.76s/it, loss=0.1661, lr=2.99e-05]
Epoch 1/8:  33%|███▎      | 956/2855 [49:17<5:08:45,  9.76s/it, loss=0.1389, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 957/2855 [49:17<3:47:16,  7.18s/it, loss=0.1389, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 957/2855 [49:18<3:47:16,  7.18s/it, loss=0.1707, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 958/2855 [49:18<2:52:49,  5.47s/it, loss=0.1707, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 958/2855 [49:22<2:52:49,  5.47s/it, loss=0.1075, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 959/2855 [49:22<2:33:56,  4.87s/it, loss=0.1075, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 959/2855 [49:24<2:33:56,  4.87s/it, loss=0.0881, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 960/2855 [49:24<2:09:10,  4.09s/it, loss=0.0881, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 960/2855 [49:26<2:09:10,  4.09s/it, loss=0.0854, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 961/2855 [49:26<1:51:29,  3.53s/it, loss=0.0854, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 961/2855 [49:28<1:51:29,  3.53s/it, loss=0.3516, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 962/2855 [49:28<1:39:21,  3.15s/it, loss=0.3516, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 962/2855 [49:31<1:39:21,  3.15s/it, loss=0.2896, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 963/2855 [49:31<1:30:32,  2.87s/it, loss=0.2896, lr=2.99e-05]
Epoch 1/8:  34%|███▎      | 963/2855 [49:33<1:30:32,  2.87s/it, loss=0.2118, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 964/2855 [49:33<1:24:36,  2.68s/it, loss=0.2118, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 964/2855 [49:35<1:24:36,  2.68s/it, loss=0.2640, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 965/2855 [49:35<1:19:08,  2.51s/it, loss=0.2640, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 965/2855 [49:37<1:19:08,  2.51s/it, loss=0.1440, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 966/2855 [49:37<1:17:15,  2.45s/it, loss=0.1440, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 966/2855 [49:40<1:17:15,  2.45s/it, loss=0.2215, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 967/2855 [49:40<1:15:27,  2.40s/it, loss=0.2215, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 967/2855 [49:42<1:15:27,  2.40s/it, loss=0.2173, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 968/2855 [49:42<1:14:49,  2.38s/it, loss=0.2173, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 968/2855 [49:44<1:14:49,  2.38s/it, loss=0.2730, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 969/2855 [49:44<1:13:08,  2.33s/it, loss=0.2730, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 969/2855 [49:47<1:13:08,  2.33s/it, loss=0.1302, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 970/2855 [49:47<1:14:29,  2.37s/it, loss=0.1302, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 970/2855 [49:49<1:14:29,  2.37s/it, loss=0.1802, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 971/2855 [49:49<1:13:30,  2.34s/it, loss=0.1802, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 971/2855 [49:51<1:13:30,  2.34s/it, loss=0.2400, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 972/2855 [49:51<1:11:52,  2.29s/it, loss=0.2400, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 972/2855 [49:53<1:11:52,  2.29s/it, loss=0.0363, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 973/2855 [49:53<1:12:04,  2.30s/it, loss=0.0363, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 973/2855 [49:56<1:12:04,  2.30s/it, loss=0.1592, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 974/2855 [49:56<1:12:39,  2.32s/it, loss=0.1592, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 974/2855 [49:58<1:12:39,  2.32s/it, loss=0.0788, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 975/2855 [49:58<1:11:36,  2.29s/it, loss=0.0788, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 975/2855 [50:00<1:11:36,  2.29s/it, loss=0.1087, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 976/2855 [50:00<1:12:26,  2.31s/it, loss=0.1087, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 976/2855 [50:03<1:12:26,  2.31s/it, loss=0.3403, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 977/2855 [50:03<1:11:34,  2.29s/it, loss=0.3403, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 977/2855 [50:05<1:11:34,  2.29s/it, loss=0.1606, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 978/2855 [50:05<1:12:24,  2.31s/it, loss=0.1606, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 978/2855 [50:08<1:12:24,  2.31s/it, loss=0.1044, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 979/2855 [50:08<1:17:08,  2.47s/it, loss=0.1044, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 979/2855 [50:10<1:17:08,  2.47s/it, loss=0.1910, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 980/2855 [50:10<1:15:30,  2.42s/it, loss=0.1910, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 980/2855 [50:12<1:15:30,  2.42s/it, loss=0.1448, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 981/2855 [50:12<1:13:51,  2.36s/it, loss=0.1448, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 981/2855 [50:15<1:13:51,  2.36s/it, loss=0.2933, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 982/2855 [50:15<1:14:36,  2.39s/it, loss=0.2933, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 982/2855 [50:17<1:14:36,  2.39s/it, loss=0.1978, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 983/2855 [50:17<1:13:19,  2.35s/it, loss=0.1978, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 983/2855 [50:19<1:13:19,  2.35s/it, loss=0.1154, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 984/2855 [50:19<1:11:44,  2.30s/it, loss=0.1154, lr=2.99e-05]
Epoch 1/8:  34%|███▍      | 984/2855 [50:22<1:11:44,  2.30s/it, loss=0.2638, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 985/2855 [50:22<1:11:29,  2.29s/it, loss=0.2638, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 985/2855 [50:24<1:11:29,  2.29s/it, loss=0.2114, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 986/2855 [50:24<1:14:15,  2.38s/it, loss=0.2114, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 986/2855 [50:27<1:14:15,  2.38s/it, loss=0.0647, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 987/2855 [50:27<1:18:17,  2.51s/it, loss=0.0647, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 987/2855 [50:29<1:18:17,  2.51s/it, loss=0.1485, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 988/2855 [50:29<1:15:03,  2.41s/it, loss=0.1485, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 988/2855 [50:31<1:15:03,  2.41s/it, loss=0.0105, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 989/2855 [50:31<1:14:20,  2.39s/it, loss=0.0105, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 989/2855 [50:34<1:14:20,  2.39s/it, loss=0.2353, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 990/2855 [50:34<1:14:52,  2.41s/it, loss=0.2353, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 990/2855 [50:36<1:14:52,  2.41s/it, loss=0.3559, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 991/2855 [50:36<1:15:08,  2.42s/it, loss=0.3559, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 991/2855 [50:43<1:15:08,  2.42s/it, loss=0.2098, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 992/2855 [50:43<1:50:55,  3.57s/it, loss=0.2098, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 992/2855 [50:45<1:50:55,  3.57s/it, loss=0.1144, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 993/2855 [50:45<1:39:38,  3.21s/it, loss=0.1144, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 993/2855 [50:48<1:39:38,  3.21s/it, loss=0.1087, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 994/2855 [50:48<1:36:01,  3.10s/it, loss=0.1087, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 994/2855 [50:50<1:36:01,  3.10s/it, loss=0.1839, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 995/2855 [50:50<1:30:47,  2.93s/it, loss=0.1839, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 995/2855 [50:53<1:30:47,  2.93s/it, loss=0.1993, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 996/2855 [50:53<1:28:20,  2.85s/it, loss=0.1993, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 996/2855 [50:56<1:28:20,  2.85s/it, loss=0.4049, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 997/2855 [50:56<1:26:20,  2.79s/it, loss=0.4049, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 997/2855 [50:59<1:26:20,  2.79s/it, loss=0.1422, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 998/2855 [50:59<1:29:00,  2.88s/it, loss=0.1422, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 998/2855 [51:01<1:29:00,  2.88s/it, loss=0.1903, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 999/2855 [51:01<1:23:00,  2.68s/it, loss=0.1903, lr=2.99e-05]
Epoch 1/8:  35%|███▍      | 999/2855 [51:03<1:23:00,  2.68s/it, loss=0.0817, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1000/2855 [51:03<1:19:19,  2.57s/it, loss=0.0817, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1000/2855 [51:06<1:19:19,  2.57s/it, loss=0.1814, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1001/2855 [51:06<1:18:59,  2.56s/it, loss=0.1814, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1001/2855 [51:08<1:18:59,  2.56s/it, loss=0.1812, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1002/2855 [51:08<1:16:42,  2.48s/it, loss=0.1812, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1002/2855 [51:10<1:16:42,  2.48s/it, loss=0.2238, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1003/2855 [51:10<1:15:15,  2.44s/it, loss=0.2238, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1003/2855 [51:13<1:15:15,  2.44s/it, loss=0.1593, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1004/2855 [51:13<1:14:39,  2.42s/it, loss=0.1593, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1004/2855 [51:15<1:14:39,  2.42s/it, loss=0.2762, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1005/2855 [51:15<1:13:47,  2.39s/it, loss=0.2762, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1005/2855 [51:17<1:13:47,  2.39s/it, loss=0.1579, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1006/2855 [51:17<1:13:03,  2.37s/it, loss=0.1579, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1006/2855 [51:20<1:13:03,  2.37s/it, loss=0.1703, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1007/2855 [51:20<1:12:05,  2.34s/it, loss=0.1703, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1007/2855 [51:22<1:12:05,  2.34s/it, loss=0.2355, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1008/2855 [51:22<1:11:09,  2.31s/it, loss=0.2355, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1008/2855 [51:24<1:11:09,  2.31s/it, loss=0.0985, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1009/2855 [51:24<1:10:34,  2.29s/it, loss=0.0985, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1009/2855 [51:27<1:10:34,  2.29s/it, loss=0.2834, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1010/2855 [51:27<1:14:13,  2.41s/it, loss=0.2834, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1010/2855 [51:29<1:14:13,  2.41s/it, loss=0.0272, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1011/2855 [51:29<1:13:20,  2.39s/it, loss=0.0272, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1011/2855 [51:31<1:13:20,  2.39s/it, loss=0.1339, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1012/2855 [51:31<1:12:06,  2.35s/it, loss=0.1339, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1012/2855 [51:34<1:12:06,  2.35s/it, loss=0.1816, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1013/2855 [51:34<1:13:15,  2.39s/it, loss=0.1816, lr=2.99e-05]
Epoch 1/8:  35%|███▌      | 1013/2855 [51:36<1:13:15,  2.39s/it, loss=0.2704, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1014/2855 [51:36<1:13:55,  2.41s/it, loss=0.2704, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1014/2855 [51:39<1:13:55,  2.41s/it, loss=0.1915, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1015/2855 [51:39<1:12:23,  2.36s/it, loss=0.1915, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1015/2855 [51:41<1:12:23,  2.36s/it, loss=0.1905, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1016/2855 [51:41<1:11:29,  2.33s/it, loss=0.1905, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1016/2855 [51:43<1:11:29,  2.33s/it, loss=0.1374, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1017/2855 [51:43<1:10:43,  2.31s/it, loss=0.1374, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1017/2855 [51:45<1:10:43,  2.31s/it, loss=0.0666, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1018/2855 [51:45<1:09:07,  2.26s/it, loss=0.0666, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1018/2855 [51:47<1:09:07,  2.26s/it, loss=0.0614, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1019/2855 [51:47<1:07:13,  2.20s/it, loss=0.0614, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1019/2855 [51:50<1:07:13,  2.20s/it, loss=0.3931, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1020/2855 [51:50<1:07:54,  2.22s/it, loss=0.3931, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1020/2855 [51:52<1:07:54,  2.22s/it, loss=0.2232, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1021/2855 [51:52<1:06:50,  2.19s/it, loss=0.2232, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1021/2855 [51:54<1:06:50,  2.19s/it, loss=0.4463, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1022/2855 [51:54<1:07:53,  2.22s/it, loss=0.4463, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1022/2855 [51:56<1:07:53,  2.22s/it, loss=0.1535, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1023/2855 [51:56<1:07:17,  2.20s/it, loss=0.1535, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1023/2855 [51:59<1:07:17,  2.20s/it, loss=0.1115, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1024/2855 [51:59<1:10:20,  2.31s/it, loss=0.1115, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1024/2855 [52:01<1:10:20,  2.31s/it, loss=0.0975, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1025/2855 [52:01<1:11:24,  2.34s/it, loss=0.0975, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1025/2855 [52:04<1:11:24,  2.34s/it, loss=0.1373, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1026/2855 [52:04<1:13:13,  2.40s/it, loss=0.1373, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1026/2855 [52:06<1:13:13,  2.40s/it, loss=0.2225, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1027/2855 [52:06<1:12:01,  2.36s/it, loss=0.2225, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1027/2855 [52:08<1:12:01,  2.36s/it, loss=0.2984, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1028/2855 [52:08<1:12:29,  2.38s/it, loss=0.2984, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1028/2855 [52:11<1:12:29,  2.38s/it, loss=0.2630, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1029/2855 [52:11<1:12:53,  2.39s/it, loss=0.2630, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1029/2855 [52:13<1:12:53,  2.39s/it, loss=0.0946, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1030/2855 [52:13<1:10:46,  2.33s/it, loss=0.0946, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1030/2855 [52:16<1:10:46,  2.33s/it, loss=0.3211, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1031/2855 [52:16<1:13:14,  2.41s/it, loss=0.3211, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1031/2855 [52:23<1:13:14,  2.41s/it, loss=0.3662, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1032/2855 [52:23<1:57:25,  3.86s/it, loss=0.3662, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1032/2855 [52:26<1:57:25,  3.86s/it, loss=0.2077, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1033/2855 [52:26<1:49:20,  3.60s/it, loss=0.2077, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1033/2855 [52:28<1:49:20,  3.60s/it, loss=0.0805, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1034/2855 [52:28<1:36:31,  3.18s/it, loss=0.0805, lr=2.99e-05]
Epoch 1/8:  36%|███▌      | 1034/2855 [52:30<1:36:31,  3.18s/it, loss=0.2652, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1035/2855 [52:30<1:28:55,  2.93s/it, loss=0.2652, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1035/2855 [52:33<1:28:55,  2.93s/it, loss=0.1487, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1036/2855 [52:33<1:23:00,  2.74s/it, loss=0.1487, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1036/2855 [52:35<1:23:00,  2.74s/it, loss=0.3699, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1037/2855 [52:35<1:18:49,  2.60s/it, loss=0.3699, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1037/2855 [52:37<1:18:49,  2.60s/it, loss=0.1754, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1038/2855 [52:37<1:17:31,  2.56s/it, loss=0.1754, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1038/2855 [52:40<1:17:31,  2.56s/it, loss=0.2478, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1039/2855 [52:40<1:13:53,  2.44s/it, loss=0.2478, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1039/2855 [52:42<1:13:53,  2.44s/it, loss=0.1506, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1040/2855 [52:42<1:13:40,  2.44s/it, loss=0.1506, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1040/2855 [52:44<1:13:40,  2.44s/it, loss=0.0478, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1041/2855 [52:44<1:12:05,  2.38s/it, loss=0.0478, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1041/2855 [52:46<1:12:05,  2.38s/it, loss=0.3118, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1042/2855 [52:46<1:10:03,  2.32s/it, loss=0.3118, lr=2.99e-05]
Epoch 1/8:  36%|███▋      | 1042/2855 [52:49<1:10:03,  2.32s/it, loss=0.2736, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1043/2855 [52:49<1:08:09,  2.26s/it, loss=0.2736, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1043/2855 [52:51<1:08:09,  2.26s/it, loss=0.3682, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1044/2855 [52:51<1:06:53,  2.22s/it, loss=0.3682, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1044/2855 [52:53<1:06:53,  2.22s/it, loss=0.2515, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1045/2855 [52:53<1:04:12,  2.13s/it, loss=0.2515, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1045/2855 [52:55<1:04:12,  2.13s/it, loss=0.2678, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1046/2855 [52:55<1:10:06,  2.33s/it, loss=0.2678, lr=2.99e-05]
Epoch 1/8:  37%|███▋      | 1046/2855 [52:58<1:10:06,  2.33s/it, loss=0.1089, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1047/2855 [52:58<1:12:22,  2.40s/it, loss=0.1089, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1047/2855 [53:05<1:12:22,  2.40s/it, loss=0.1700, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1048/2855 [53:05<1:49:31,  3.64s/it, loss=0.1700, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1048/2855 [53:08<1:49:31,  3.64s/it, loss=0.1473, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1049/2855 [53:08<1:48:16,  3.60s/it, loss=0.1473, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1049/2855 [53:10<1:48:16,  3.60s/it, loss=0.2520, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1050/2855 [53:10<1:35:38,  3.18s/it, loss=0.2520, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1050/2855 [53:13<1:35:38,  3.18s/it, loss=0.2632, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1051/2855 [53:13<1:31:24,  3.04s/it, loss=0.2632, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1051/2855 [53:15<1:31:24,  3.04s/it, loss=0.0873, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1052/2855 [53:15<1:24:38,  2.82s/it, loss=0.0873, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1052/2855 [53:18<1:24:38,  2.82s/it, loss=0.2187, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1053/2855 [53:18<1:20:47,  2.69s/it, loss=0.2187, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1053/2855 [53:20<1:20:47,  2.69s/it, loss=0.1193, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1054/2855 [53:20<1:19:01,  2.63s/it, loss=0.1193, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1054/2855 [53:22<1:19:01,  2.63s/it, loss=0.1868, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1055/2855 [53:22<1:15:32,  2.52s/it, loss=0.1868, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1055/2855 [53:25<1:15:32,  2.52s/it, loss=0.1134, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1056/2855 [53:25<1:15:38,  2.52s/it, loss=0.1134, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1056/2855 [53:27<1:15:38,  2.52s/it, loss=0.0735, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1057/2855 [53:27<1:15:02,  2.50s/it, loss=0.0735, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1057/2855 [53:30<1:15:02,  2.50s/it, loss=0.2615, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1058/2855 [53:30<1:14:20,  2.48s/it, loss=0.2615, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1058/2855 [53:32<1:14:20,  2.48s/it, loss=0.1166, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1059/2855 [53:32<1:13:10,  2.44s/it, loss=0.1166, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1059/2855 [53:34<1:13:10,  2.44s/it, loss=0.1053, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1060/2855 [53:34<1:11:30,  2.39s/it, loss=0.1053, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1060/2855 [53:37<1:11:30,  2.39s/it, loss=0.3098, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1061/2855 [53:37<1:11:34,  2.39s/it, loss=0.3098, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1061/2855 [53:39<1:11:34,  2.39s/it, loss=0.1882, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1062/2855 [53:39<1:10:48,  2.37s/it, loss=0.1882, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1062/2855 [53:41<1:10:48,  2.37s/it, loss=0.3890, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1063/2855 [53:41<1:10:06,  2.35s/it, loss=0.3890, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1063/2855 [53:44<1:10:06,  2.35s/it, loss=0.1875, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1064/2855 [53:44<1:09:18,  2.32s/it, loss=0.1875, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1064/2855 [53:46<1:09:18,  2.32s/it, loss=0.2363, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1065/2855 [53:46<1:08:17,  2.29s/it, loss=0.2363, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1065/2855 [53:48<1:08:17,  2.29s/it, loss=0.2759, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1066/2855 [53:48<1:10:37,  2.37s/it, loss=0.2759, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1066/2855 [53:51<1:10:37,  2.37s/it, loss=0.0599, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1067/2855 [53:51<1:11:45,  2.41s/it, loss=0.0599, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1067/2855 [53:54<1:11:45,  2.41s/it, loss=0.1227, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1068/2855 [53:54<1:14:42,  2.51s/it, loss=0.1227, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1068/2855 [54:00<1:14:42,  2.51s/it, loss=0.2071, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1069/2855 [54:00<1:46:19,  3.57s/it, loss=0.2071, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1069/2855 [54:03<1:46:19,  3.57s/it, loss=0.1753, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1070/2855 [54:03<1:44:23,  3.51s/it, loss=0.1753, lr=2.98e-05]
Epoch 1/8:  37%|███▋      | 1070/2855 [54:11<1:44:23,  3.51s/it, loss=0.1438, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1071/2855 [54:11<2:21:57,  4.77s/it, loss=0.1438, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1071/2855 [54:13<2:21:57,  4.77s/it, loss=0.0055, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1072/2855 [54:13<1:54:13,  3.84s/it, loss=0.0055, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1072/2855 [54:16<1:54:13,  3.84s/it, loss=0.2124, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1073/2855 [54:16<1:47:10,  3.61s/it, loss=0.2124, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1073/2855 [54:18<1:47:10,  3.61s/it, loss=0.3537, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1074/2855 [54:18<1:32:59,  3.13s/it, loss=0.3537, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1074/2855 [54:20<1:32:59,  3.13s/it, loss=0.2696, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1075/2855 [54:20<1:27:37,  2.95s/it, loss=0.2696, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1075/2855 [54:23<1:27:37,  2.95s/it, loss=0.1359, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1076/2855 [54:23<1:23:19,  2.81s/it, loss=0.1359, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1076/2855 [54:25<1:23:19,  2.81s/it, loss=0.1281, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1077/2855 [54:25<1:19:12,  2.67s/it, loss=0.1281, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1077/2855 [54:27<1:19:12,  2.67s/it, loss=0.1748, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1078/2855 [54:27<1:14:09,  2.50s/it, loss=0.1748, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1078/2855 [54:30<1:14:09,  2.50s/it, loss=0.0809, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1079/2855 [54:30<1:13:34,  2.49s/it, loss=0.0809, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1079/2855 [54:32<1:13:34,  2.49s/it, loss=0.0408, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1080/2855 [54:32<1:12:41,  2.46s/it, loss=0.0408, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1080/2855 [54:34<1:12:41,  2.46s/it, loss=0.1602, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1081/2855 [54:34<1:12:38,  2.46s/it, loss=0.1602, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1081/2855 [54:37<1:12:38,  2.46s/it, loss=0.1441, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1082/2855 [54:37<1:11:42,  2.43s/it, loss=0.1441, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1082/2855 [54:39<1:11:42,  2.43s/it, loss=0.2262, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1083/2855 [54:39<1:09:53,  2.37s/it, loss=0.2262, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1083/2855 [54:41<1:09:53,  2.37s/it, loss=0.2025, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1084/2855 [54:41<1:09:20,  2.35s/it, loss=0.2025, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1084/2855 [54:44<1:09:20,  2.35s/it, loss=0.3159, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1085/2855 [54:44<1:09:00,  2.34s/it, loss=0.3159, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1085/2855 [54:46<1:09:00,  2.34s/it, loss=0.2349, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1086/2855 [54:46<1:07:47,  2.30s/it, loss=0.2349, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1086/2855 [54:48<1:07:47,  2.30s/it, loss=0.2984, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1087/2855 [54:48<1:06:31,  2.26s/it, loss=0.2984, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1087/2855 [54:50<1:06:31,  2.26s/it, loss=0.4184, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1088/2855 [54:50<1:08:12,  2.32s/it, loss=0.4184, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1088/2855 [54:53<1:08:12,  2.32s/it, loss=0.2115, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1089/2855 [54:53<1:07:19,  2.29s/it, loss=0.2115, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1089/2855 [54:55<1:07:19,  2.29s/it, loss=0.2145, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1090/2855 [54:55<1:07:52,  2.31s/it, loss=0.2145, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1090/2855 [54:57<1:07:52,  2.31s/it, loss=0.2031, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1091/2855 [54:57<1:07:17,  2.29s/it, loss=0.2031, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1091/2855 [55:00<1:07:17,  2.29s/it, loss=0.0479, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1092/2855 [55:00<1:07:34,  2.30s/it, loss=0.0479, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1092/2855 [55:02<1:07:34,  2.30s/it, loss=0.3217, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1093/2855 [55:02<1:06:14,  2.26s/it, loss=0.3217, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1093/2855 [55:04<1:06:14,  2.26s/it, loss=0.1962, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1094/2855 [55:04<1:07:14,  2.29s/it, loss=0.1962, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1094/2855 [55:06<1:07:14,  2.29s/it, loss=0.1874, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1095/2855 [55:06<1:07:30,  2.30s/it, loss=0.1874, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1095/2855 [55:09<1:07:30,  2.30s/it, loss=0.2707, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1096/2855 [55:09<1:06:39,  2.27s/it, loss=0.2707, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1096/2855 [55:11<1:06:39,  2.27s/it, loss=0.0961, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1097/2855 [55:11<1:05:57,  2.25s/it, loss=0.0961, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1097/2855 [55:13<1:05:57,  2.25s/it, loss=0.1503, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1098/2855 [55:13<1:05:17,  2.23s/it, loss=0.1503, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1098/2855 [55:15<1:05:17,  2.23s/it, loss=0.1530, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1099/2855 [55:15<1:06:17,  2.26s/it, loss=0.1530, lr=2.98e-05]
Epoch 1/8:  38%|███▊      | 1099/2855 [55:17<1:06:17,  2.26s/it, loss=0.3550, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1100/2855 [55:17<1:04:37,  2.21s/it, loss=0.3550, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1100/2855 [55:20<1:04:37,  2.21s/it, loss=0.2692, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1101/2855 [55:20<1:06:54,  2.29s/it, loss=0.2692, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1101/2855 [55:22<1:06:54,  2.29s/it, loss=0.1727, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1102/2855 [55:22<1:06:29,  2.28s/it, loss=0.1727, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1102/2855 [55:24<1:06:29,  2.28s/it, loss=0.0500, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1103/2855 [55:24<1:06:15,  2.27s/it, loss=0.0500, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1103/2855 [55:27<1:06:15,  2.27s/it, loss=0.1458, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1104/2855 [55:27<1:05:14,  2.24s/it, loss=0.1458, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1104/2855 [55:29<1:05:14,  2.24s/it, loss=0.2333, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1105/2855 [55:29<1:05:43,  2.25s/it, loss=0.2333, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1105/2855 [55:31<1:05:43,  2.25s/it, loss=0.1729, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1106/2855 [55:31<1:06:02,  2.27s/it, loss=0.1729, lr=2.98e-05]
Epoch 1/8:  39%|███▊      | 1106/2855 [55:33<1:06:02,  2.27s/it, loss=0.2203, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1107/2855 [55:33<1:03:35,  2.18s/it, loss=0.2203, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1107/2855 [55:35<1:03:35,  2.18s/it, loss=0.1045, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1108/2855 [55:35<1:03:08,  2.17s/it, loss=0.1045, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1108/2855 [55:37<1:03:08,  2.17s/it, loss=0.1339, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1109/2855 [55:37<1:01:32,  2.11s/it, loss=0.1339, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1109/2855 [55:39<1:01:32,  2.11s/it, loss=0.2257, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1110/2855 [55:39<1:02:18,  2.14s/it, loss=0.2257, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1110/2855 [55:42<1:02:18,  2.14s/it, loss=0.0848, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1111/2855 [55:42<1:06:51,  2.30s/it, loss=0.0848, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1111/2855 [55:44<1:06:51,  2.30s/it, loss=0.2302, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1112/2855 [55:44<1:02:35,  2.15s/it, loss=0.2302, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1112/2855 [55:46<1:02:35,  2.15s/it, loss=0.2545, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1113/2855 [55:46<1:03:40,  2.19s/it, loss=0.2545, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1113/2855 [55:48<1:03:40,  2.19s/it, loss=0.0675, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1114/2855 [55:48<1:00:29,  2.08s/it, loss=0.0675, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1114/2855 [55:50<1:00:29,  2.08s/it, loss=0.0388, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1115/2855 [55:50<1:01:33,  2.12s/it, loss=0.0388, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1115/2855 [55:52<1:01:33,  2.12s/it, loss=0.3099, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1116/2855 [55:52<1:00:29,  2.09s/it, loss=0.3099, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1116/2855 [55:55<1:00:29,  2.09s/it, loss=0.1668, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1117/2855 [55:55<1:03:57,  2.21s/it, loss=0.1668, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1117/2855 [55:57<1:03:57,  2.21s/it, loss=0.1928, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1118/2855 [55:57<1:04:09,  2.22s/it, loss=0.1928, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1118/2855 [55:59<1:04:09,  2.22s/it, loss=0.1265, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1119/2855 [55:59<1:05:07,  2.25s/it, loss=0.1265, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1119/2855 [56:02<1:05:07,  2.25s/it, loss=0.2068, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1120/2855 [56:02<1:05:33,  2.27s/it, loss=0.2068, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1120/2855 [56:04<1:05:33,  2.27s/it, loss=0.2722, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1121/2855 [56:04<1:03:09,  2.19s/it, loss=0.2722, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1121/2855 [56:06<1:03:09,  2.19s/it, loss=0.1210, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1122/2855 [56:06<1:03:47,  2.21s/it, loss=0.1210, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1122/2855 [56:08<1:03:47,  2.21s/it, loss=0.2688, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1123/2855 [56:08<1:02:38,  2.17s/it, loss=0.2688, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1123/2855 [56:10<1:02:38,  2.17s/it, loss=0.3128, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1124/2855 [56:10<1:03:07,  2.19s/it, loss=0.3128, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1124/2855 [56:12<1:03:07,  2.19s/it, loss=0.2165, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1125/2855 [56:12<1:02:44,  2.18s/it, loss=0.2165, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1125/2855 [56:15<1:02:44,  2.18s/it, loss=0.3066, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1126/2855 [56:15<1:04:05,  2.22s/it, loss=0.3066, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1126/2855 [56:17<1:04:05,  2.22s/it, loss=0.1556, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1127/2855 [56:17<1:04:34,  2.24s/it, loss=0.1556, lr=2.98e-05]
Epoch 1/8:  39%|███▉      | 1127/2855 [56:19<1:04:34,  2.24s/it, loss=0.1065, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1128/2855 [56:19<1:03:08,  2.19s/it, loss=0.1065, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1128/2855 [56:21<1:03:08,  2.19s/it, loss=0.1481, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1129/2855 [56:21<1:04:08,  2.23s/it, loss=0.1481, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1129/2855 [56:23<1:04:08,  2.23s/it, loss=0.2473, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1130/2855 [56:23<1:03:13,  2.20s/it, loss=0.2473, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1130/2855 [56:26<1:03:13,  2.20s/it, loss=0.1450, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1131/2855 [56:26<1:04:51,  2.26s/it, loss=0.1450, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1131/2855 [56:28<1:04:51,  2.26s/it, loss=0.2516, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1132/2855 [56:28<1:01:59,  2.16s/it, loss=0.2516, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1132/2855 [56:31<1:01:59,  2.16s/it, loss=0.1920, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1133/2855 [56:31<1:09:33,  2.42s/it, loss=0.1920, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1133/2855 [56:33<1:09:33,  2.42s/it, loss=0.2683, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1134/2855 [56:33<1:07:52,  2.37s/it, loss=0.2683, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1134/2855 [56:35<1:07:52,  2.37s/it, loss=0.1033, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1135/2855 [56:35<1:07:39,  2.36s/it, loss=0.1033, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1135/2855 [56:38<1:07:39,  2.36s/it, loss=0.3529, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1136/2855 [56:38<1:07:36,  2.36s/it, loss=0.3529, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1136/2855 [56:40<1:07:36,  2.36s/it, loss=0.1913, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1137/2855 [56:40<1:07:18,  2.35s/it, loss=0.1913, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1137/2855 [56:42<1:07:18,  2.35s/it, loss=0.2094, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1138/2855 [56:42<1:06:41,  2.33s/it, loss=0.2094, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1138/2855 [56:45<1:06:41,  2.33s/it, loss=0.4092, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1139/2855 [56:45<1:07:28,  2.36s/it, loss=0.4092, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1139/2855 [56:47<1:07:28,  2.36s/it, loss=0.2457, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1140/2855 [56:47<1:06:20,  2.32s/it, loss=0.2457, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1140/2855 [56:49<1:06:20,  2.32s/it, loss=0.3506, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1141/2855 [56:49<1:04:50,  2.27s/it, loss=0.3506, lr=2.98e-05]
Epoch 1/8:  40%|███▉      | 1141/2855 [56:52<1:04:50,  2.27s/it, loss=0.3234, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1142/2855 [56:52<1:05:32,  2.30s/it, loss=0.3234, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1142/2855 [56:54<1:05:32,  2.30s/it, loss=0.3594, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1143/2855 [56:54<1:05:37,  2.30s/it, loss=0.3594, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1143/2855 [56:56<1:05:37,  2.30s/it, loss=0.2481, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1144/2855 [56:56<1:07:11,  2.36s/it, loss=0.2481, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1144/2855 [56:59<1:07:11,  2.36s/it, loss=0.2179, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1145/2855 [56:59<1:06:22,  2.33s/it, loss=0.2179, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1145/2855 [57:01<1:06:22,  2.33s/it, loss=0.1714, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1146/2855 [57:01<1:06:05,  2.32s/it, loss=0.1714, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1146/2855 [57:03<1:06:05,  2.32s/it, loss=0.0916, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1147/2855 [57:03<1:04:05,  2.25s/it, loss=0.0916, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1147/2855 [57:05<1:04:05,  2.25s/it, loss=0.2036, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1148/2855 [57:05<1:01:53,  2.18s/it, loss=0.2036, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1148/2855 [57:07<1:01:53,  2.18s/it, loss=0.1015, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1149/2855 [57:07<1:01:18,  2.16s/it, loss=0.1015, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1149/2855 [57:10<1:01:18,  2.16s/it, loss=0.2095, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1150/2855 [57:10<1:05:41,  2.31s/it, loss=0.2095, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1150/2855 [57:12<1:05:41,  2.31s/it, loss=0.1063, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1151/2855 [57:12<1:06:29,  2.34s/it, loss=0.1063, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1151/2855 [57:15<1:06:29,  2.34s/it, loss=0.2065, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1152/2855 [57:15<1:11:34,  2.52s/it, loss=0.2065, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1152/2855 [57:18<1:11:34,  2.52s/it, loss=0.1827, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1153/2855 [57:18<1:13:18,  2.58s/it, loss=0.1827, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1153/2855 [57:20<1:13:18,  2.58s/it, loss=0.0700, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1154/2855 [57:20<1:11:20,  2.52s/it, loss=0.0700, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1154/2855 [57:22<1:11:20,  2.52s/it, loss=0.3217, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1155/2855 [57:22<1:08:57,  2.43s/it, loss=0.3217, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1155/2855 [57:25<1:08:57,  2.43s/it, loss=0.1345, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1156/2855 [57:25<1:07:36,  2.39s/it, loss=0.1345, lr=2.98e-05]
Epoch 1/8:  40%|████      | 1156/2855 [57:27<1:07:36,  2.39s/it, loss=0.1390, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1157/2855 [57:27<1:06:04,  2.33s/it, loss=0.1390, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1157/2855 [57:30<1:06:04,  2.33s/it, loss=0.0896, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1158/2855 [57:30<1:09:27,  2.46s/it, loss=0.0896, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1158/2855 [57:32<1:09:27,  2.46s/it, loss=0.2178, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1159/2855 [57:32<1:11:13,  2.52s/it, loss=0.2178, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1159/2855 [57:36<1:11:13,  2.52s/it, loss=0.0721, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1160/2855 [57:36<1:19:15,  2.81s/it, loss=0.0721, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1160/2855 [59:39<1:19:15,  2.81s/it, loss=0.1530, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1161/2855 [59:39<18:16:21, 38.83s/it, loss=0.1530, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1161/2855 [59:41<18:16:21, 38.83s/it, loss=0.0982, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1162/2855 [59:41<13:05:45, 27.85s/it, loss=0.0982, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1162/2855 [59:42<13:05:45, 27.85s/it, loss=0.0076, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1163/2855 [59:42<9:21:02, 19.89s/it, loss=0.0076, lr=2.98e-05] 
Epoch 1/8:  41%|████      | 1163/2855 [59:44<9:21:02, 19.89s/it, loss=0.1366, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1164/2855 [59:44<6:50:12, 14.55s/it, loss=0.1366, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1164/2855 [59:47<6:50:12, 14.55s/it, loss=0.3420, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1165/2855 [59:47<5:10:51, 11.04s/it, loss=0.3420, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1165/2855 [59:49<5:10:51, 11.04s/it, loss=0.3226, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1166/2855 [59:49<3:52:58,  8.28s/it, loss=0.3226, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1166/2855 [59:51<3:52:58,  8.28s/it, loss=0.0995, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1167/2855 [59:51<3:00:56,  6.43s/it, loss=0.0995, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1167/2855 [59:53<3:00:56,  6.43s/it, loss=0.2619, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1168/2855 [59:53<2:23:57,  5.12s/it, loss=0.2619, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1168/2855 [59:56<2:23:57,  5.12s/it, loss=0.1732, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1169/2855 [59:56<2:01:08,  4.31s/it, loss=0.1732, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1169/2855 [59:57<2:01:08,  4.31s/it, loss=0.0965, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1170/2855 [59:57<1:38:51,  3.52s/it, loss=0.0965, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1170/2855 [59:59<1:38:51,  3.52s/it, loss=0.2818, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1171/2855 [59:59<1:24:36,  3.01s/it, loss=0.2818, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1171/2855 [1:00:01<1:24:36,  3.01s/it, loss=0.2102, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1172/2855 [1:00:01<1:17:18,  2.76s/it, loss=0.2102, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1172/2855 [1:00:03<1:17:18,  2.76s/it, loss=0.1866, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1173/2855 [1:00:03<1:08:35,  2.45s/it, loss=0.1866, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1173/2855 [1:00:05<1:08:35,  2.45s/it, loss=0.1852, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1174/2855 [1:00:05<1:06:09,  2.36s/it, loss=0.1852, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1174/2855 [1:00:07<1:06:09,  2.36s/it, loss=0.1630, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1175/2855 [1:00:07<1:00:46,  2.17s/it, loss=0.1630, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1175/2855 [1:00:09<1:00:46,  2.17s/it, loss=0.2823, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1176/2855 [1:00:09<57:06,  2.04s/it, loss=0.2823, lr=2.98e-05]  
Epoch 1/8:  41%|████      | 1176/2855 [1:00:11<57:06,  2.04s/it, loss=0.3802, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1177/2855 [1:00:11<58:54,  2.11s/it, loss=0.3802, lr=2.98e-05]
Epoch 1/8:  41%|████      | 1177/2855 [1:00:13<58:54,  2.11s/it, loss=0.1896, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1178/2855 [1:00:13<56:08,  2.01s/it, loss=0.1896, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1178/2855 [1:00:15<56:08,  2.01s/it, loss=0.1813, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1179/2855 [1:00:15<58:03,  2.08s/it, loss=0.1813, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1179/2855 [1:00:17<58:03,  2.08s/it, loss=0.1626, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1180/2855 [1:00:17<57:27,  2.06s/it, loss=0.1626, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1180/2855 [1:00:19<57:27,  2.06s/it, loss=0.2211, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1181/2855 [1:00:19<56:01,  2.01s/it, loss=0.2211, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1181/2855 [1:00:21<56:01,  2.01s/it, loss=0.2052, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1182/2855 [1:00:21<57:11,  2.05s/it, loss=0.2052, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1182/2855 [1:00:23<57:11,  2.05s/it, loss=0.0913, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1183/2855 [1:00:23<53:34,  1.92s/it, loss=0.0913, lr=2.98e-05]
Epoch 1/8:  41%|███���▏     | 1183/2855 [1:00:24<53:34,  1.92s/it, loss=0.2148, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1184/2855 [1:00:24<52:52,  1.90s/it, loss=0.2148, lr=2.98e-05]
Epoch 1/8:  41%|████▏     | 1184/2855 [1:00:26<52:52,  1.90s/it, loss=0.2358, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1185/2855 [1:00:26<53:19,  1.92s/it, loss=0.2358, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1185/2855 [1:00:28<53:19,  1.92s/it, loss=0.2610, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1186/2855 [1:00:28<50:36,  1.82s/it, loss=0.2610, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1186/2855 [1:00:30<50:36,  1.82s/it, loss=0.1705, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1187/2855 [1:00:30<54:27,  1.96s/it, loss=0.1705, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1187/2855 [1:00:32<54:27,  1.96s/it, loss=0.2643, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1188/2855 [1:00:32<54:30,  1.96s/it, loss=0.2643, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1188/2855 [1:00:34<54:30,  1.96s/it, loss=0.3558, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1189/2855 [1:00:34<54:22,  1.96s/it, loss=0.3558, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1189/2855 [1:00:38<54:22,  1.96s/it, loss=0.2489, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1190/2855 [1:00:38<1:10:26,  2.54s/it, loss=0.2489, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1190/2855 [1:00:41<1:10:26,  2.54s/it, loss=0.3389, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1191/2855 [1:00:41<1:10:19,  2.54s/it, loss=0.3389, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1191/2855 [1:00:42<1:10:19,  2.54s/it, loss=0.1109, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1192/2855 [1:00:42<1:03:43,  2.30s/it, loss=0.1109, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1192/2855 [1:00:44<1:03:43,  2.30s/it, loss=0.1936, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1193/2855 [1:00:44<58:23,  2.11s/it, loss=0.1936, lr=2.98e-05]  
Epoch 1/8:  42%|████▏     | 1193/2855 [1:00:46<58:23,  2.11s/it, loss=0.1104, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1194/2855 [1:00:46<58:46,  2.12s/it, loss=0.1104, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1194/2855 [1:00:48<58:46,  2.12s/it, loss=0.1338, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1195/2855 [1:00:48<55:05,  1.99s/it, loss=0.1338, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1195/2855 [1:00:50<55:05,  1.99s/it, loss=0.3447, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1196/2855 [1:00:50<54:12,  1.96s/it, loss=0.3447, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1196/2855 [1:00:52<54:12,  1.96s/it, loss=0.2603, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1197/2855 [1:00:52<54:13,  1.96s/it, loss=0.2603, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1197/2855 [1:00:53<54:13,  1.96s/it, loss=0.3788, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1198/2855 [1:00:53<51:16,  1.86s/it, loss=0.3788, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1198/2855 [1:00:55<51:16,  1.86s/it, loss=0.1036, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1199/2855 [1:00:55<52:43,  1.91s/it, loss=0.1036, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1199/2855 [1:00:57<52:43,  1.91s/it, loss=0.3139, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1200/2855 [1:00:57<50:57,  1.85s/it, loss=0.3139, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1200/2855 [1:00:59<50:57,  1.85s/it, loss=0.2951, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1201/2855 [1:00:59<49:34,  1.80s/it, loss=0.2951, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1201/2855 [1:01:01<49:34,  1.80s/it, loss=0.3146, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1202/2855 [1:01:01<56:50,  2.06s/it, loss=0.3146, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1202/2855 [1:01:04<56:50,  2.06s/it, loss=0.2388, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1203/2855 [1:01:04<56:44,  2.06s/it, loss=0.2388, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1203/2855 [1:01:06<56:44,  2.06s/it, loss=0.1762, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1204/2855 [1:01:06<59:05,  2.15s/it, loss=0.1762, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1204/2855 [1:01:08<59:05,  2.15s/it, loss=0.2951, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1205/2855 [1:01:08<56:14,  2.05s/it, loss=0.2951, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1205/2855 [1:01:09<56:14,  2.05s/it, loss=0.1504, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1206/2855 [1:01:09<53:32,  1.95s/it, loss=0.1504, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1206/2855 [1:01:12<53:32,  1.95s/it, loss=0.2448, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1207/2855 [1:01:12<54:58,  2.00s/it, loss=0.2448, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1207/2855 [1:01:13<54:58,  2.00s/it, loss=0.0456, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1208/2855 [1:01:13<51:47,  1.89s/it, loss=0.0456, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1208/2855 [1:01:15<51:47,  1.89s/it, loss=0.2665, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1209/2855 [1:01:15<51:22,  1.87s/it, loss=0.2665, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1209/2855 [1:01:17<51:22,  1.87s/it, loss=0.2810, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1210/2855 [1:01:17<50:50,  1.85s/it, loss=0.2810, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1210/2855 [1:01:18<50:50,  1.85s/it, loss=0.5777, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1211/2855 [1:01:18<48:06,  1.76s/it, loss=0.5777, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1211/2855 [1:01:20<48:06,  1.76s/it, loss=0.4265, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1212/2855 [1:01:20<48:04,  1.76s/it, loss=0.4265, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1212/2855 [1:01:22<48:04,  1.76s/it, loss=0.3193, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1213/2855 [1:01:22<47:45,  1.75s/it, loss=0.3193, lr=2.98e-05]
Epoch 1/8:  42%|████▏     | 1213/2855 [1:01:23<47:45,  1.75s/it, loss=0.1676, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1214/2855 [1:01:23<46:02,  1.68s/it, loss=0.1676, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1214/2855 [1:01:25<46:02,  1.68s/it, loss=0.3741, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1215/2855 [1:01:25<48:25,  1.77s/it, loss=0.3741, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1215/2855 [1:01:27<48:25,  1.77s/it, loss=0.2235, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1216/2855 [1:01:27<48:52,  1.79s/it, loss=0.2235, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1216/2855 [1:01:29<48:52,  1.79s/it, loss=0.0812, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1217/2855 [1:01:29<48:29,  1.78s/it, loss=0.0812, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1217/2855 [1:01:31<48:29,  1.78s/it, loss=0.1209, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1218/2855 [1:01:31<51:25,  1.88s/it, loss=0.1209, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1218/2855 [1:01:33<51:25,  1.88s/it, loss=0.1028, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1219/2855 [1:01:33<51:07,  1.87s/it, loss=0.1028, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1219/2855 [1:01:35<51:07,  1.87s/it, loss=0.1427, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1220/2855 [1:01:35<53:17,  1.96s/it, loss=0.1427, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1220/2855 [1:01:37<53:17,  1.96s/it, loss=0.2103, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1221/2855 [1:01:37<55:03,  2.02s/it, loss=0.2103, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1221/2855 [1:01:39<55:03,  2.02s/it, loss=0.2761, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1222/2855 [1:01:39<52:07,  1.92s/it, loss=0.2761, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1222/2855 [1:01:41<52:07,  1.92s/it, loss=0.3047, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1223/2855 [1:01:41<53:50,  1.98s/it, loss=0.3047, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1223/2855 [1:01:43<53:50,  1.98s/it, loss=0.3297, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1224/2855 [1:01:43<52:21,  1.93s/it, loss=0.3297, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1224/2855 [1:01:45<52:21,  1.93s/it, loss=0.2824, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1225/2855 [1:01:45<51:46,  1.91s/it, loss=0.2824, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1225/2855 [1:01:47<51:46,  1.91s/it, loss=0.1939, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1226/2855 [1:01:47<54:14,  2.00s/it, loss=0.1939, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1226/2855 [1:01:49<54:14,  2.00s/it, loss=0.0757, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1227/2855 [1:01:49<51:29,  1.90s/it, loss=0.0757, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1227/2855 [1:01:51<51:29,  1.90s/it, loss=0.2008, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1228/2855 [1:01:51<54:51,  2.02s/it, loss=0.2008, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1228/2855 [1:01:53<54:51,  2.02s/it, loss=0.0470, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1229/2855 [1:01:53<53:25,  1.97s/it, loss=0.0470, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1229/2855 [1:01:55<53:25,  1.97s/it, loss=0.1330, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1230/2855 [1:01:55<54:52,  2.03s/it, loss=0.1330, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1230/2855 [1:01:57<54:52,  2.03s/it, loss=0.1303, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1231/2855 [1:01:57<54:14,  2.00s/it, loss=0.1303, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1231/2855 [1:01:58<54:14,  2.00s/it, loss=0.2481, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1232/2855 [1:01:58<50:36,  1.87s/it, loss=0.2481, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1232/2855 [1:02:00<50:36,  1.87s/it, loss=0.1976, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1233/2855 [1:02:00<49:09,  1.82s/it, loss=0.1976, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1233/2855 [1:02:02<49:09,  1.82s/it, loss=0.0646, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1234/2855 [1:02:02<51:17,  1.90s/it, loss=0.0646, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1234/2855 [1:02:04<51:17,  1.90s/it, loss=0.1764, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1235/2855 [1:02:04<48:29,  1.80s/it, loss=0.1764, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1235/2855 [1:02:05<48:29,  1.80s/it, loss=0.0543, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1236/2855 [1:02:05<47:25,  1.76s/it, loss=0.0543, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1236/2855 [1:02:07<47:25,  1.76s/it, loss=0.0273, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1237/2855 [1:02:07<48:12,  1.79s/it, loss=0.0273, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1237/2855 [1:02:09<48:12,  1.79s/it, loss=0.1644, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1238/2855 [1:02:09<46:20,  1.72s/it, loss=0.1644, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1238/2855 [1:02:11<46:20,  1.72s/it, loss=0.1664, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1239/2855 [1:02:11<48:43,  1.81s/it, loss=0.1664, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1239/2855 [1:02:13<48:43,  1.81s/it, loss=0.1778, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1240/2855 [1:02:13<49:15,  1.83s/it, loss=0.1778, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1240/2855 [1:02:14<49:15,  1.83s/it, loss=0.2846, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1241/2855 [1:02:14<46:55,  1.74s/it, loss=0.2846, lr=2.98e-05]
Epoch 1/8:  43%|████▎     | 1241/2855 [1:02:16<46:55,  1.74s/it, loss=0.0313, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1242/2855 [1:02:16<48:24,  1.80s/it, loss=0.0313, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1242/2855 [1:02:18<48:24,  1.80s/it, loss=0.1561, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1243/2855 [1:02:18<48:04,  1.79s/it, loss=0.1561, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1243/2855 [1:02:20<48:04,  1.79s/it, loss=0.0107, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1244/2855 [1:02:20<46:55,  1.75s/it, loss=0.0107, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1244/2855 [1:02:22<46:55,  1.75s/it, loss=0.0877, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1245/2855 [1:02:22<48:37,  1.81s/it, loss=0.0877, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1245/2855 [1:02:23<48:37,  1.81s/it, loss=0.3765, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1246/2855 [1:02:23<46:59,  1.75s/it, loss=0.3765, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1246/2855 [1:02:25<46:59,  1.75s/it, loss=0.0765, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1247/2855 [1:02:25<46:34,  1.74s/it, loss=0.0765, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1247/2855 [1:02:27<46:34,  1.74s/it, loss=0.1650, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1248/2855 [1:02:27<49:01,  1.83s/it, loss=0.1650, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1248/2855 [1:02:28<49:01,  1.83s/it, loss=0.2899, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1249/2855 [1:02:28<46:37,  1.74s/it, loss=0.2899, lr=2.98e-05]
Epoch 1/8:  44%|████▎     | 1249/2855 [1:02:30<46:37,  1.74s/it, loss=0.2631, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1250/2855 [1:02:30<45:28,  1.70s/it, loss=0.2631, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1250/2855 [1:02:32<45:28,  1.70s/it, loss=0.1595, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1251/2855 [1:02:32<47:20,  1.77s/it, loss=0.1595, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1251/2855 [1:02:34<47:20,  1.77s/it, loss=0.2657, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1252/2855 [1:02:34<45:23,  1.70s/it, loss=0.2657, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1252/2855 [1:02:35<45:23,  1.70s/it, loss=0.1721, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1253/2855 [1:02:35<44:44,  1.68s/it, loss=0.1721, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1253/2855 [1:02:37<44:44,  1.68s/it, loss=0.4348, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1254/2855 [1:02:37<47:10,  1.77s/it, loss=0.4348, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1254/2855 [1:02:39<47:10,  1.77s/it, loss=0.2564, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1255/2855 [1:02:39<45:08,  1.69s/it, loss=0.2564, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1255/2855 [1:02:40<45:08,  1.69s/it, loss=0.0893, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1256/2855 [1:02:40<45:05,  1.69s/it, loss=0.0893, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1256/2855 [1:02:42<45:05,  1.69s/it, loss=0.2161, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1257/2855 [1:02:42<47:53,  1.80s/it, loss=0.2161, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1257/2855 [1:02:44<47:53,  1.80s/it, loss=0.2396, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1258/2855 [1:02:44<45:31,  1.71s/it, loss=0.2396, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1258/2855 [1:02:46<45:31,  1.71s/it, loss=0.1477, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1259/2855 [1:02:46<46:16,  1.74s/it, loss=0.1477, lr=2.98e-05]
Epoch 1/8:  44%|█���██▍     | 1259/2855 [1:02:48<46:16,  1.74s/it, loss=0.2940, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1260/2855 [1:02:48<47:23,  1.78s/it, loss=0.2940, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1260/2855 [1:02:49<47:23,  1.78s/it, loss=0.1371, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1261/2855 [1:02:49<45:34,  1.72s/it, loss=0.1371, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1261/2855 [1:02:51<45:34,  1.72s/it, loss=0.1944, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1262/2855 [1:02:51<47:15,  1.78s/it, loss=0.1944, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1262/2855 [1:02:53<47:15,  1.78s/it, loss=0.1287, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1263/2855 [1:02:53<47:09,  1.78s/it, loss=0.1287, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1263/2855 [1:02:55<47:09,  1.78s/it, loss=0.2593, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1264/2855 [1:02:55<50:14,  1.89s/it, loss=0.2593, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1264/2855 [1:02:57<50:14,  1.89s/it, loss=0.1302, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1265/2855 [1:02:57<54:24,  2.05s/it, loss=0.1302, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1265/2855 [1:02:59<54:24,  2.05s/it, loss=0.1418, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1266/2855 [1:02:59<51:08,  1.93s/it, loss=0.1418, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1266/2855 [1:03:01<51:08,  1.93s/it, loss=0.2199, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1267/2855 [1:03:01<50:12,  1.90s/it, loss=0.2199, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1267/2855 [1:03:03<50:12,  1.90s/it, loss=0.2882, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1268/2855 [1:03:03<51:34,  1.95s/it, loss=0.2882, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1268/2855 [1:03:05<51:34,  1.95s/it, loss=0.1001, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1269/2855 [1:03:05<51:12,  1.94s/it, loss=0.1001, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1269/2855 [1:03:07<51:12,  1.94s/it, loss=0.2018, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1270/2855 [1:03:07<53:24,  2.02s/it, loss=0.2018, lr=2.98e-05]
Epoch 1/8:  44%|████▍     | 1270/2855 [1:03:09<53:24,  2.02s/it, loss=0.2325, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1271/2855 [1:03:09<51:18,  1.94s/it, loss=0.2325, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1271/2855 [1:03:11<51:18,  1.94s/it, loss=0.0797, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1272/2855 [1:03:11<50:52,  1.93s/it, loss=0.0797, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1272/2855 [1:03:13<50:52,  1.93s/it, loss=0.1503, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1273/2855 [1:03:13<52:29,  1.99s/it, loss=0.1503, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1273/2855 [1:03:15<52:29,  1.99s/it, loss=0.1098, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1274/2855 [1:03:15<50:32,  1.92s/it, loss=0.1098, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1274/2855 [1:03:17<50:32,  1.92s/it, loss=0.2038, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1275/2855 [1:03:17<52:30,  1.99s/it, loss=0.2038, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1275/2855 [1:03:18<52:30,  1.99s/it, loss=0.1697, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1276/2855 [1:03:18<49:40,  1.89s/it, loss=0.1697, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1276/2855 [1:03:20<49:40,  1.89s/it, loss=0.1190, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1277/2855 [1:03:20<46:42,  1.78s/it, loss=0.1190, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1277/2855 [1:03:22<46:42,  1.78s/it, loss=0.4388, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1278/2855 [1:03:22<48:13,  1.84s/it, loss=0.4388, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1278/2855 [1:03:24<48:13,  1.84s/it, loss=0.3134, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1279/2855 [1:03:24<46:21,  1.77s/it, loss=0.3134, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1279/2855 [1:03:25<46:21,  1.77s/it, loss=0.2346, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1280/2855 [1:03:25<44:49,  1.71s/it, loss=0.2346, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1280/2855 [1:03:27<44:49,  1.71s/it, loss=0.0977, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1281/2855 [1:03:27<48:19,  1.84s/it, loss=0.0977, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1281/2855 [1:03:29<48:19,  1.84s/it, loss=0.1520, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1282/2855 [1:03:29<46:23,  1.77s/it, loss=0.1520, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1282/2855 [1:03:31<46:23,  1.77s/it, loss=0.2023, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1283/2855 [1:03:31<45:17,  1.73s/it, loss=0.2023, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1283/2855 [1:03:33<45:17,  1.73s/it, loss=0.8166, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1284/2855 [1:03:33<47:42,  1.82s/it, loss=0.8166, lr=2.98e-05]
Epoch 1/8:  45%|████▍     | 1284/2855 [1:03:34<47:42,  1.82s/it, loss=0.2146, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1285/2855 [1:03:34<45:21,  1.73s/it, loss=0.2146, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1285/2855 [1:03:36<45:21,  1.73s/it, loss=0.3600, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1286/2855 [1:03:36<45:17,  1.73s/it, loss=0.3600, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1286/2855 [1:03:38<45:17,  1.73s/it, loss=0.1807, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1287/2855 [1:03:38<49:34,  1.90s/it, loss=0.1807, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1287/2855 [1:03:40<49:34,  1.90s/it, loss=0.1660, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1288/2855 [1:03:40<46:59,  1.80s/it, loss=0.1660, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1288/2855 [1:03:42<46:59,  1.80s/it, loss=0.1589, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1289/2855 [1:03:42<49:38,  1.90s/it, loss=0.1589, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1289/2855 [1:03:44<49:38,  1.90s/it, loss=0.1101, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1290/2855 [1:03:44<48:45,  1.87s/it, loss=0.1101, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1290/2855 [1:03:45<48:45,  1.87s/it, loss=0.1284, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1291/2855 [1:03:45<48:16,  1.85s/it, loss=0.1284, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1291/2855 [1:03:48<48:16,  1.85s/it, loss=0.2999, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1292/2855 [1:03:48<53:18,  2.05s/it, loss=0.2999, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1292/2855 [1:03:50<53:18,  2.05s/it, loss=0.2011, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1293/2855 [1:03:50<51:50,  1.99s/it, loss=0.2011, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1293/2855 [1:03:52<51:50,  1.99s/it, loss=0.2586, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1294/2855 [1:03:52<52:57,  2.04s/it, loss=0.2586, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1294/2855 [1:03:54<52:57,  2.04s/it, loss=0.2224, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1295/2855 [1:03:54<52:03,  2.00s/it, loss=0.2224, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1295/2855 [1:03:56<52:03,  2.00s/it, loss=0.2160, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1296/2855 [1:03:56<50:52,  1.96s/it, loss=0.2160, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1296/2855 [1:03:58<50:52,  1.96s/it, loss=0.1288, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1297/2855 [1:03:58<53:45,  2.07s/it, loss=0.1288, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1297/2855 [1:04:00<53:45,  2.07s/it, loss=0.2107, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1298/2855 [1:04:00<52:38,  2.03s/it, loss=0.2107, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1298/2855 [1:04:02<52:38,  2.03s/it, loss=0.0231, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1299/2855 [1:04:02<53:46,  2.07s/it, loss=0.0231, lr=2.98e-05]
Epoch 1/8:  45%|████▌     | 1299/2855 [1:04:04<53:46,  2.07s/it, loss=0.1934, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1300/2855 [1:04:04<52:07,  2.01s/it, loss=0.1934, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1300/2855 [1:04:06<52:07,  2.01s/it, loss=0.2220, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1301/2855 [1:04:06<50:13,  1.94s/it, loss=0.2220, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1301/2855 [1:04:08<50:13,  1.94s/it, loss=0.2085, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1302/2855 [1:04:08<53:00,  2.05s/it, loss=0.2085, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1302/2855 [1:04:10<53:00,  2.05s/it, loss=0.2812, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1303/2855 [1:04:10<50:54,  1.97s/it, loss=0.2812, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1303/2855 [1:04:12<50:54,  1.97s/it, loss=0.1695, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1304/2855 [1:04:12<51:16,  1.98s/it, loss=0.1695, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1304/2855 [1:04:14<51:16,  1.98s/it, loss=0.1742, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1305/2855 [1:04:14<49:43,  1.92s/it, loss=0.1742, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1305/2855 [1:04:15<49:43,  1.92s/it, loss=0.2620, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1306/2855 [1:04:15<47:07,  1.83s/it, loss=0.2620, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1306/2855 [1:04:17<47:07,  1.83s/it, loss=0.1004, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1307/2855 [1:04:17<49:26,  1.92s/it, loss=0.1004, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1307/2855 [1:04:19<49:26,  1.92s/it, loss=0.1771, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1308/2855 [1:04:19<47:16,  1.83s/it, loss=0.1771, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1308/2855 [1:04:21<47:16,  1.83s/it, loss=0.3005, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1309/2855 [1:04:21<46:44,  1.81s/it, loss=0.3005, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1309/2855 [1:04:23<46:44,  1.81s/it, loss=0.2183, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1310/2855 [1:04:23<49:18,  1.91s/it, loss=0.2183, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1310/2855 [1:04:25<49:18,  1.91s/it, loss=0.3531, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1311/2855 [1:04:25<47:25,  1.84s/it, loss=0.3531, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1311/2855 [1:04:26<47:25,  1.84s/it, loss=0.1055, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1312/2855 [1:04:26<47:25,  1.84s/it, loss=0.1055, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1312/2855 [1:04:29<47:25,  1.84s/it, loss=0.1756, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1313/2855 [1:04:29<49:15,  1.92s/it, loss=0.1756, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1313/2855 [1:04:30<49:15,  1.92s/it, loss=0.1976, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1314/2855 [1:04:30<46:34,  1.81s/it, loss=0.1976, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1314/2855 [1:04:32<46:34,  1.81s/it, loss=0.3114, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1315/2855 [1:04:32<49:14,  1.92s/it, loss=0.3114, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1315/2855 [1:04:34<49:14,  1.92s/it, loss=0.1410, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1316/2855 [1:04:34<47:40,  1.86s/it, loss=0.1410, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1316/2855 [1:04:36<47:40,  1.86s/it, loss=0.2572, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1317/2855 [1:04:36<45:17,  1.77s/it, loss=0.2572, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1317/2855 [1:04:38<45:17,  1.77s/it, loss=0.2847, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1318/2855 [1:04:38<49:23,  1.93s/it, loss=0.2847, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1318/2855 [1:04:40<49:23,  1.93s/it, loss=0.3854, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1319/2855 [1:04:40<48:02,  1.88s/it, loss=0.3854, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1319/2855 [1:04:42<48:02,  1.88s/it, loss=0.2015, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1320/2855 [1:04:42<49:47,  1.95s/it, loss=0.2015, lr=2.98e-05]
Epoch 1/8:  46%|████▌     | 1320/2855 [1:04:44<49:47,  1.95s/it, loss=0.5265, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1321/2855 [1:04:44<53:16,  2.08s/it, loss=0.5265, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1321/2855 [1:04:46<53:16,  2.08s/it, loss=0.1652, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1322/2855 [1:04:46<51:03,  2.00s/it, loss=0.1652, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1322/2855 [1:04:48<51:03,  2.00s/it, loss=0.1878, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1323/2855 [1:04:48<52:27,  2.05s/it, loss=0.1878, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1323/2855 [1:04:50<52:27,  2.05s/it, loss=0.1806, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1324/2855 [1:04:50<50:21,  1.97s/it, loss=0.1806, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1324/2855 [1:04:52<50:21,  1.97s/it, loss=0.2323, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1325/2855 [1:04:52<50:17,  1.97s/it, loss=0.2323, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1325/2855 [1:04:54<50:17,  1.97s/it, loss=0.3450, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1326/2855 [1:04:54<52:35,  2.06s/it, loss=0.3450, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1326/2855 [1:04:56<52:35,  2.06s/it, loss=0.2151, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1327/2855 [1:04:56<50:45,  1.99s/it, loss=0.2151, lr=2.98e-05]
Epoch 1/8:  46%|████▋     | 1327/2855 [1:04:58<50:45,  1.99s/it, loss=0.1682, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1328/2855 [1:04:58<52:33,  2.07s/it, loss=0.1682, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1328/2855 [1:05:00<52:33,  2.07s/it, loss=0.2144, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1329/2855 [1:05:00<51:17,  2.02s/it, loss=0.2144, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1329/2855 [1:05:02<51:17,  2.02s/it, loss=0.1359, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1330/2855 [1:05:02<50:28,  1.99s/it, loss=0.1359, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1330/2855 [1:05:05<50:28,  1.99s/it, loss=0.2325, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1331/2855 [1:05:05<54:32,  2.15s/it, loss=0.2325, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1331/2855 [1:05:06<54:32,  2.15s/it, loss=0.3896, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1332/2855 [1:05:06<51:37,  2.03s/it, loss=0.3896, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1332/2855 [1:05:08<51:37,  2.03s/it, loss=0.4026, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1333/2855 [1:05:08<52:03,  2.05s/it, loss=0.4026, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1333/2855 [1:05:10<52:03,  2.05s/it, loss=0.2612, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1334/2855 [1:05:10<49:48,  1.97s/it, loss=0.2612, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1334/2855 [1:05:12<49:48,  1.97s/it, loss=0.1041, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1335/2855 [1:05:12<50:25,  1.99s/it, loss=0.1041, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1335/2855 [1:05:14<50:25,  1.99s/it, loss=0.0685, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1336/2855 [1:05:14<49:45,  1.97s/it, loss=0.0685, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1336/2855 [1:05:16<49:45,  1.97s/it, loss=0.1552, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1337/2855 [1:05:16<46:13,  1.83s/it, loss=0.1552, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1337/2855 [1:05:18<46:13,  1.83s/it, loss=0.3079, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1338/2855 [1:05:18<48:03,  1.90s/it, loss=0.3079, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1338/2855 [1:05:20<48:03,  1.90s/it, loss=0.2157, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1339/2855 [1:05:20<48:49,  1.93s/it, loss=0.2157, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1339/2855 [1:05:21<48:49,  1.93s/it, loss=0.0920, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1340/2855 [1:05:21<46:51,  1.86s/it, loss=0.0920, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1340/2855 [1:05:24<46:51,  1.86s/it, loss=0.3122, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1341/2855 [1:05:24<49:38,  1.97s/it, loss=0.3122, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1341/2855 [1:05:25<49:38,  1.97s/it, loss=0.1725, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1342/2855 [1:05:25<46:42,  1.85s/it, loss=0.1725, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1342/2855 [1:05:27<46:42,  1.85s/it, loss=0.1232, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1343/2855 [1:05:27<45:50,  1.82s/it, loss=0.1232, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1343/2855 [1:05:29<45:50,  1.82s/it, loss=0.0162, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1344/2855 [1:05:29<48:03,  1.91s/it, loss=0.0162, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1344/2855 [1:05:31<48:03,  1.91s/it, loss=0.2726, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1345/2855 [1:05:31<45:02,  1.79s/it, loss=0.2726, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1345/2855 [1:05:32<45:02,  1.79s/it, loss=0.1762, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1346/2855 [1:05:32<45:55,  1.83s/it, loss=0.1762, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1346/2855 [1:05:34<45:55,  1.83s/it, loss=0.1909, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1347/2855 [1:05:34<47:04,  1.87s/it, loss=0.1909, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1347/2855 [1:05:36<47:04,  1.87s/it, loss=0.4445, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1348/2855 [1:05:36<46:12,  1.84s/it, loss=0.4445, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1348/2855 [1:05:39<46:12,  1.84s/it, loss=0.2146, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1349/2855 [1:05:39<51:57,  2.07s/it, loss=0.2146, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1349/2855 [1:05:41<51:57,  2.07s/it, loss=0.1884, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1350/2855 [1:05:41<50:09,  2.00s/it, loss=0.1884, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1350/2855 [1:05:43<50:09,  2.00s/it, loss=0.2148, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1351/2855 [1:05:43<49:55,  1.99s/it, loss=0.2148, lr=2.98e-05]
Epoch 1/8:  47%|████▋     | 1351/2855 [1:05:45<49:55,  1.99s/it, loss=0.2232, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1352/2855 [1:05:45<51:49,  2.07s/it, loss=0.2232, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1352/2855 [1:05:47<51:49,  2.07s/it, loss=0.2742, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1353/2855 [1:05:47<50:05,  2.00s/it, loss=0.2742, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1353/2855 [1:05:49<50:05,  2.00s/it, loss=0.0821, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1354/2855 [1:05:49<51:52,  2.07s/it, loss=0.0821, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1354/2855 [1:05:51<51:52,  2.07s/it, loss=0.3984, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1355/2855 [1:05:51<49:53,  2.00s/it, loss=0.3984, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1355/2855 [1:05:53<49:53,  2.00s/it, loss=0.3522, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1356/2855 [1:05:53<50:36,  2.03s/it, loss=0.3522, lr=2.97e-05]
Epoch 1/8:  47%|████▋     | 1356/2855 [1:05:55<50:36,  2.03s/it, loss=0.0932, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1357/2855 [1:05:55<51:37,  2.07s/it, loss=0.0932, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1357/2855 [1:05:57<51:37,  2.07s/it, loss=0.4067, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1358/2855 [1:05:57<49:35,  1.99s/it, loss=0.4067, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1358/2855 [1:05:59<49:35,  1.99s/it, loss=0.3117, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1359/2855 [1:05:59<51:15,  2.06s/it, loss=0.3117, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1359/2855 [1:06:01<51:15,  2.06s/it, loss=0.1595, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1360/2855 [1:06:01<49:15,  1.98s/it, loss=0.1595, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1360/2855 [1:06:03<49:15,  1.98s/it, loss=0.1612, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1361/2855 [1:06:03<49:36,  1.99s/it, loss=0.1612, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1361/2855 [1:06:05<49:36,  1.99s/it, loss=0.3497, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1362/2855 [1:06:05<50:07,  2.01s/it, loss=0.3497, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1362/2855 [1:06:07<50:07,  2.01s/it, loss=0.1227, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1363/2855 [1:06:07<46:53,  1.89s/it, loss=0.1227, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1363/2855 [1:06:09<46:53,  1.89s/it, loss=0.0610, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1364/2855 [1:06:09<50:20,  2.03s/it, loss=0.0610, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1364/2855 [1:06:11<50:20,  2.03s/it, loss=0.3618, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1365/2855 [1:06:11<48:36,  1.96s/it, loss=0.3618, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1365/2855 [1:06:12<48:36,  1.96s/it, loss=0.1521, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1366/2855 [1:06:12<47:06,  1.90s/it, loss=0.1521, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1366/2855 [1:06:15<47:06,  1.90s/it, loss=0.0426, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1367/2855 [1:06:15<48:46,  1.97s/it, loss=0.0426, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1367/2855 [1:06:16<48:46,  1.97s/it, loss=0.1369, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1368/2855 [1:06:16<46:18,  1.87s/it, loss=0.1369, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1368/2855 [1:06:18<46:18,  1.87s/it, loss=0.1494, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1369/2855 [1:06:18<46:32,  1.88s/it, loss=0.1494, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1369/2855 [1:06:20<46:32,  1.88s/it, loss=0.0992, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1370/2855 [1:06:20<47:30,  1.92s/it, loss=0.0992, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1370/2855 [1:06:22<47:30,  1.92s/it, loss=0.2922, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1371/2855 [1:06:22<44:28,  1.80s/it, loss=0.2922, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1371/2855 [1:06:24<44:28,  1.80s/it, loss=0.1221, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1372/2855 [1:06:24<46:11,  1.87s/it, loss=0.1221, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1372/2855 [1:06:26<46:11,  1.87s/it, loss=0.2151, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1373/2855 [1:06:26<46:02,  1.86s/it, loss=0.2151, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1373/2855 [1:06:27<46:02,  1.86s/it, loss=0.1580, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1374/2855 [1:06:27<44:57,  1.82s/it, loss=0.1580, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1374/2855 [1:06:29<44:57,  1.82s/it, loss=0.2529, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1375/2855 [1:06:29<47:08,  1.91s/it, loss=0.2529, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1375/2855 [1:06:31<47:08,  1.91s/it, loss=0.1742, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1376/2855 [1:06:31<44:38,  1.81s/it, loss=0.1742, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1376/2855 [1:06:33<44:38,  1.81s/it, loss=0.2513, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1377/2855 [1:06:33<44:10,  1.79s/it, loss=0.2513, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1377/2855 [1:06:35<44:10,  1.79s/it, loss=0.1362, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1378/2855 [1:06:35<46:09,  1.88s/it, loss=0.1362, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1378/2855 [1:06:36<46:09,  1.88s/it, loss=0.2766, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1379/2855 [1:06:36<44:13,  1.80s/it, loss=0.2766, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1379/2855 [1:06:38<44:13,  1.80s/it, loss=0.2719, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1380/2855 [1:06:38<45:32,  1.85s/it, loss=0.2719, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1380/2855 [1:06:40<45:32,  1.85s/it, loss=0.1188, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1381/2855 [1:06:40<46:53,  1.91s/it, loss=0.1188, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1381/2855 [1:06:42<46:53,  1.91s/it, loss=0.3033, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1382/2855 [1:06:42<44:46,  1.82s/it, loss=0.3033, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1382/2855 [1:06:44<44:46,  1.82s/it, loss=0.1674, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1383/2855 [1:06:44<47:53,  1.95s/it, loss=0.1674, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1383/2855 [1:06:46<47:53,  1.95s/it, loss=0.0250, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1384/2855 [1:06:46<49:29,  2.02s/it, loss=0.0250, lr=2.97e-05]
Epoch 1/8:  48%|████▊     | 1384/2855 [1:06:48<49:29,  2.02s/it, loss=0.2535, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1385/2855 [1:06:48<48:44,  1.99s/it, loss=0.2535, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1385/2855 [1:06:50<48:44,  1.99s/it, loss=0.0889, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1386/2855 [1:06:50<49:26,  2.02s/it, loss=0.0889, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1386/2855 [1:06:52<49:26,  2.02s/it, loss=0.0274, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1387/2855 [1:06:52<47:41,  1.95s/it, loss=0.0274, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1387/2855 [1:06:55<47:41,  1.95s/it, loss=0.3311, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1388/2855 [1:06:55<50:12,  2.05s/it, loss=0.3311, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1388/2855 [1:06:57<50:12,  2.05s/it, loss=0.2249, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1389/2855 [1:06:57<49:31,  2.03s/it, loss=0.2249, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1389/2855 [1:06:58<49:31,  2.03s/it, loss=0.2902, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1390/2855 [1:06:58<48:33,  1.99s/it, loss=0.2902, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1390/2855 [1:07:01<48:33,  1.99s/it, loss=0.1784, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1391/2855 [1:07:01<49:41,  2.04s/it, loss=0.1784, lr=2.97e-05]
Epoch 1/8:  49%|████▊     | 1391/2855 [1:07:02<49:41,  2.04s/it, loss=0.3485, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1392/2855 [1:07:02<47:37,  1.95s/it, loss=0.3485, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1392/2855 [1:07:04<47:37,  1.95s/it, loss=0.1807, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1393/2855 [1:07:04<48:07,  1.97s/it, loss=0.1807, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1393/2855 [1:07:06<48:07,  1.97s/it, loss=0.2236, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1394/2855 [1:07:06<48:40,  2.00s/it, loss=0.2236, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1394/2855 [1:07:08<48:40,  2.00s/it, loss=0.0956, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1395/2855 [1:07:08<47:22,  1.95s/it, loss=0.0956, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1395/2855 [1:07:10<47:22,  1.95s/it, loss=0.0245, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1396/2855 [1:07:10<47:53,  1.97s/it, loss=0.0245, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1396/2855 [1:07:12<47:53,  1.97s/it, loss=0.2498, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1397/2855 [1:07:12<45:35,  1.88s/it, loss=0.2498, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1397/2855 [1:07:14<45:35,  1.88s/it, loss=0.4033, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1398/2855 [1:07:14<47:26,  1.95s/it, loss=0.4033, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1398/2855 [1:07:16<47:26,  1.95s/it, loss=0.1309, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1399/2855 [1:07:16<46:57,  1.94s/it, loss=0.1309, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1399/2855 [1:07:18<46:57,  1.94s/it, loss=0.1839, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1400/2855 [1:07:18<44:50,  1.85s/it, loss=0.1839, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1400/2855 [1:07:20<44:50,  1.85s/it, loss=0.2510, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1401/2855 [1:07:20<47:43,  1.97s/it, loss=0.2510, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1401/2855 [1:07:22<47:43,  1.97s/it, loss=0.1088, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1402/2855 [1:07:22<48:15,  1.99s/it, loss=0.1088, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1402/2855 [1:07:24<48:15,  1.99s/it, loss=0.2201, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1403/2855 [1:07:24<48:28,  2.00s/it, loss=0.2201, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1403/2855 [1:07:26<48:28,  2.00s/it, loss=0.3107, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1404/2855 [1:07:26<49:33,  2.05s/it, loss=0.3107, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1404/2855 [1:07:28<49:33,  2.05s/it, loss=0.2603, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1405/2855 [1:07:28<47:39,  1.97s/it, loss=0.2603, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1405/2855 [1:07:30<47:39,  1.97s/it, loss=0.3245, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1406/2855 [1:07:30<49:01,  2.03s/it, loss=0.3245, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1406/2855 [1:07:32<49:01,  2.03s/it, loss=0.0850, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1407/2855 [1:07:32<48:01,  1.99s/it, loss=0.0850, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1407/2855 [1:07:34<48:01,  1.99s/it, loss=0.3443, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1408/2855 [1:07:34<47:49,  1.98s/it, loss=0.3443, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1408/2855 [1:07:36<47:49,  1.98s/it, loss=0.1301, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1409/2855 [1:07:36<49:10,  2.04s/it, loss=0.1301, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1409/2855 [1:07:38<49:10,  2.04s/it, loss=0.3104, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1410/2855 [1:07:38<47:23,  1.97s/it, loss=0.3104, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1410/2855 [1:07:40<47:23,  1.97s/it, loss=0.1890, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1411/2855 [1:07:40<48:43,  2.02s/it, loss=0.1890, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1411/2855 [1:07:42<48:43,  2.02s/it, loss=0.2734, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1412/2855 [1:07:42<48:15,  2.01s/it, loss=0.2734, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1412/2855 [1:07:44<48:15,  2.01s/it, loss=0.0546, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1413/2855 [1:07:44<46:22,  1.93s/it, loss=0.0546, lr=2.97e-05]
Epoch 1/8:  49%|████▉     | 1413/2855 [1:07:46<46:22,  1.93s/it, loss=0.1783, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1414/2855 [1:07:46<46:24,  1.93s/it, loss=0.1783, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1414/2855 [1:07:47<46:24,  1.93s/it, loss=0.1684, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1415/2855 [1:07:47<43:23,  1.81s/it, loss=0.1684, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1415/2855 [1:07:49<43:23,  1.81s/it, loss=0.3036, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1416/2855 [1:07:49<44:03,  1.84s/it, loss=0.3036, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1416/2855 [1:07:51<44:03,  1.84s/it, loss=0.1612, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1417/2855 [1:07:51<45:31,  1.90s/it, loss=0.1612, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1417/2855 [1:07:53<45:31,  1.90s/it, loss=0.1968, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1418/2855 [1:07:53<44:52,  1.87s/it, loss=0.1968, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1418/2855 [1:07:56<44:52,  1.87s/it, loss=0.2089, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1419/2855 [1:07:56<49:51,  2.08s/it, loss=0.2089, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1419/2855 [1:07:57<49:51,  2.08s/it, loss=0.2200, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1420/2855 [1:07:57<47:57,  2.01s/it, loss=0.2200, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1420/2855 [1:07:59<47:57,  2.01s/it, loss=0.2285, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1421/2855 [1:07:59<47:04,  1.97s/it, loss=0.2285, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1421/2855 [1:08:01<47:04,  1.97s/it, loss=0.1086, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1422/2855 [1:08:01<48:37,  2.04s/it, loss=0.1086, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1422/2855 [1:08:03<48:37,  2.04s/it, loss=0.1125, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1423/2855 [1:08:03<47:06,  1.97s/it, loss=0.1125, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1423/2855 [1:08:06<47:06,  1.97s/it, loss=0.2494, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1424/2855 [1:08:06<49:42,  2.08s/it, loss=0.2494, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1424/2855 [1:08:07<49:42,  2.08s/it, loss=0.1106, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1425/2855 [1:08:07<48:04,  2.02s/it, loss=0.1106, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1425/2855 [1:08:09<48:04,  2.02s/it, loss=0.1667, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1426/2855 [1:08:09<47:31,  2.00s/it, loss=0.1667, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1426/2855 [1:08:12<47:31,  2.00s/it, loss=0.1290, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1427/2855 [1:08:12<49:32,  2.08s/it, loss=0.1290, lr=2.97e-05]
Epoch 1/8:  50%|████▉     | 1427/2855 [1:08:14<49:32,  2.08s/it, loss=0.1054, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1428/2855 [1:08:14<47:56,  2.02s/it, loss=0.1054, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1428/2855 [1:08:16<47:56,  2.02s/it, loss=0.2655, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1429/2855 [1:08:16<49:54,  2.10s/it, loss=0.2655, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1429/2855 [1:08:18<49:54,  2.10s/it, loss=0.0890, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1430/2855 [1:08:18<47:06,  1.98s/it, loss=0.0890, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1430/2855 [1:08:19<47:06,  1.98s/it, loss=0.3034, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1431/2855 [1:08:19<46:19,  1.95s/it, loss=0.3034, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1431/2855 [1:08:21<46:19,  1.95s/it, loss=0.1639, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1432/2855 [1:08:21<45:37,  1.92s/it, loss=0.1639, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1432/2855 [1:08:23<45:37,  1.92s/it, loss=0.3325, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1433/2855 [1:08:23<43:33,  1.84s/it, loss=0.3325, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1433/2855 [1:08:25<43:33,  1.84s/it, loss=0.1035, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1434/2855 [1:08:25<46:40,  1.97s/it, loss=0.1035, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1434/2855 [1:08:27<46:40,  1.97s/it, loss=0.2170, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1435/2855 [1:08:27<44:48,  1.89s/it, loss=0.2170, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1435/2855 [1:08:29<44:48,  1.89s/it, loss=0.1973, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1436/2855 [1:08:29<43:37,  1.84s/it, loss=0.1973, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1436/2855 [1:08:31<43:37,  1.84s/it, loss=0.1434, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1437/2855 [1:08:31<46:19,  1.96s/it, loss=0.1434, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1437/2855 [1:08:33<46:19,  1.96s/it, loss=0.2105, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1438/2855 [1:08:33<43:52,  1.86s/it, loss=0.2105, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1438/2855 [1:08:34<43:52,  1.86s/it, loss=0.2174, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1439/2855 [1:08:34<43:51,  1.86s/it, loss=0.2174, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1439/2855 [1:08:37<43:51,  1.86s/it, loss=0.1714, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1440/2855 [1:08:37<46:51,  1.99s/it, loss=0.1714, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1440/2855 [1:08:39<46:51,  1.99s/it, loss=0.1054, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1441/2855 [1:08:39<47:46,  2.03s/it, loss=0.1054, lr=2.97e-05]
Epoch 1/8:  50%|█████     | 1441/2855 [1:08:41<47:46,  2.03s/it, loss=0.3503, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1442/2855 [1:08:41<50:11,  2.13s/it, loss=0.3503, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1442/2855 [1:08:43<50:11,  2.13s/it, loss=0.1449, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1443/2855 [1:08:43<47:39,  2.03s/it, loss=0.1449, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1443/2855 [1:08:45<47:39,  2.03s/it, loss=0.4053, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1444/2855 [1:08:45<47:32,  2.02s/it, loss=0.4053, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1444/2855 [1:08:47<47:32,  2.02s/it, loss=0.2162, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1445/2855 [1:08:47<48:08,  2.05s/it, loss=0.2162, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1445/2855 [1:08:49<48:08,  2.05s/it, loss=0.1078, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1446/2855 [1:08:49<46:45,  1.99s/it, loss=0.1078, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1446/2855 [1:08:51<46:45,  1.99s/it, loss=0.1485, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1447/2855 [1:08:51<48:51,  2.08s/it, loss=0.1485, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1447/2855 [1:08:53<48:51,  2.08s/it, loss=0.0923, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1448/2855 [1:08:53<47:18,  2.02s/it, loss=0.0923, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1448/2855 [1:08:55<47:18,  2.02s/it, loss=0.2327, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1449/2855 [1:08:55<49:49,  2.13s/it, loss=0.2327, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1449/2855 [1:08:57<49:49,  2.13s/it, loss=0.4208, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1450/2855 [1:08:57<48:29,  2.07s/it, loss=0.4208, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1450/2855 [1:08:59<48:29,  2.07s/it, loss=0.2694, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1451/2855 [1:08:59<46:37,  1.99s/it, loss=0.2694, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1451/2855 [1:09:01<46:37,  1.99s/it, loss=0.1380, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1452/2855 [1:09:01<48:08,  2.06s/it, loss=0.1380, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1452/2855 [1:09:03<48:08,  2.06s/it, loss=0.3363, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1453/2855 [1:09:03<45:58,  1.97s/it, loss=0.3363, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1453/2855 [1:09:05<45:58,  1.97s/it, loss=0.1575, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1454/2855 [1:09:05<46:12,  1.98s/it, loss=0.1575, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1454/2855 [1:09:07<46:12,  1.98s/it, loss=0.2537, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1455/2855 [1:09:07<44:56,  1.93s/it, loss=0.2537, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1455/2855 [1:09:09<44:56,  1.93s/it, loss=0.2638, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1456/2855 [1:09:09<42:49,  1.84s/it, loss=0.2638, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1456/2855 [1:09:11<42:49,  1.84s/it, loss=0.0418, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1457/2855 [1:09:11<45:29,  1.95s/it, loss=0.0418, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1457/2855 [1:09:12<45:29,  1.95s/it, loss=0.1258, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1458/2855 [1:09:12<43:27,  1.87s/it, loss=0.1258, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1458/2855 [1:09:14<43:27,  1.87s/it, loss=0.0513, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1459/2855 [1:09:14<43:39,  1.88s/it, loss=0.0513, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1459/2855 [1:09:17<43:39,  1.88s/it, loss=0.2127, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1460/2855 [1:09:17<48:18,  2.08s/it, loss=0.2127, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1460/2855 [1:09:19<48:18,  2.08s/it, loss=0.1763, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1461/2855 [1:09:19<47:00,  2.02s/it, loss=0.1763, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1461/2855 [1:09:21<47:00,  2.02s/it, loss=0.1658, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1462/2855 [1:09:21<50:17,  2.17s/it, loss=0.1658, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1462/2855 [1:09:23<50:17,  2.17s/it, loss=0.2480, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1463/2855 [1:09:23<47:39,  2.05s/it, loss=0.2480, lr=2.97e-05]
Epoch 1/8:  51%|█████     | 1463/2855 [1:09:25<47:39,  2.05s/it, loss=0.1153, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1464/2855 [1:09:25<47:34,  2.05s/it, loss=0.1153, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1464/2855 [1:09:27<47:34,  2.05s/it, loss=0.2318, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1465/2855 [1:09:27<48:04,  2.08s/it, loss=0.2318, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1465/2855 [1:09:29<48:04,  2.08s/it, loss=0.2668, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1466/2855 [1:09:29<46:13,  2.00s/it, loss=0.2668, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1466/2855 [1:09:31<46:13,  2.00s/it, loss=0.1683, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1467/2855 [1:09:31<48:33,  2.10s/it, loss=0.1683, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1467/2855 [1:09:33<48:33,  2.10s/it, loss=0.0821, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1468/2855 [1:09:33<46:23,  2.01s/it, loss=0.0821, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1468/2855 [1:09:35<46:23,  2.01s/it, loss=0.3648, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1469/2855 [1:09:35<47:18,  2.05s/it, loss=0.3648, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1469/2855 [1:09:38<47:18,  2.05s/it, loss=0.1815, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1470/2855 [1:09:38<48:28,  2.10s/it, loss=0.1815, lr=2.97e-05]
Epoch 1/8:  51%|█████▏    | 1470/2855 [1:09:39<48:28,  2.10s/it, loss=0.0398, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1471/2855 [1:09:39<45:56,  1.99s/it, loss=0.0398, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1471/2855 [1:09:41<45:56,  1.99s/it, loss=0.2234, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1472/2855 [1:09:41<46:05,  2.00s/it, loss=0.2234, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1472/2855 [1:09:43<46:05,  2.00s/it, loss=0.2416, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1473/2855 [1:09:43<43:27,  1.89s/it, loss=0.2416, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1473/2855 [1:09:45<43:27,  1.89s/it, loss=0.4399, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1474/2855 [1:09:45<42:43,  1.86s/it, loss=0.4399, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1474/2855 [1:09:47<42:43,  1.86s/it, loss=0.0850, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1475/2855 [1:09:47<45:09,  1.96s/it, loss=0.0850, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1475/2855 [1:09:49<45:09,  1.96s/it, loss=0.1832, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1476/2855 [1:09:49<42:21,  1.84s/it, loss=0.1832, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1476/2855 [1:09:51<42:21,  1.84s/it, loss=0.2847, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1477/2855 [1:09:51<47:38,  2.07s/it, loss=0.2847, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1477/2855 [1:09:53<47:38,  2.07s/it, loss=0.0722, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1478/2855 [1:09:53<46:37,  2.03s/it, loss=0.0722, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1478/2855 [1:09:55<46:37,  2.03s/it, loss=0.2620, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1479/2855 [1:09:55<45:23,  1.98s/it, loss=0.2620, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1479/2855 [1:09:57<45:23,  1.98s/it, loss=0.2075, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1480/2855 [1:09:57<48:15,  2.11s/it, loss=0.2075, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1480/2855 [1:09:59<48:15,  2.11s/it, loss=0.0938, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1481/2855 [1:09:59<46:23,  2.03s/it, loss=0.0938, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1481/2855 [1:10:01<46:23,  2.03s/it, loss=0.1128, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1482/2855 [1:10:01<47:13,  2.06s/it, loss=0.1128, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1482/2855 [1:10:03<47:13,  2.06s/it, loss=0.1621, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1483/2855 [1:10:03<45:43,  2.00s/it, loss=0.1621, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1483/2855 [1:10:05<45:43,  2.00s/it, loss=0.3662, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1484/2855 [1:10:05<45:08,  1.98s/it, loss=0.3662, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1484/2855 [1:10:07<45:08,  1.98s/it, loss=0.0939, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1485/2855 [1:10:08<47:54,  2.10s/it, loss=0.0939, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1485/2855 [1:10:10<47:54,  2.10s/it, loss=0.2156, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1486/2855 [1:10:10<47:15,  2.07s/it, loss=0.2156, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1486/2855 [1:10:12<47:15,  2.07s/it, loss=0.4455, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1487/2855 [1:10:12<49:04,  2.15s/it, loss=0.4455, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1487/2855 [1:10:14<49:04,  2.15s/it, loss=0.1829, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1488/2855 [1:10:14<47:51,  2.10s/it, loss=0.1829, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1488/2855 [1:10:16<47:51,  2.10s/it, loss=0.1493, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1489/2855 [1:10:16<46:33,  2.05s/it, loss=0.1493, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1489/2855 [1:10:18<46:33,  2.05s/it, loss=0.1192, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1490/2855 [1:10:18<47:27,  2.09s/it, loss=0.1192, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1490/2855 [1:10:20<47:27,  2.09s/it, loss=0.1756, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1491/2855 [1:10:20<44:42,  1.97s/it, loss=0.1756, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1491/2855 [1:10:22<44:42,  1.97s/it, loss=0.1590, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1492/2855 [1:10:22<46:44,  2.06s/it, loss=0.1590, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1492/2855 [1:10:24<46:44,  2.06s/it, loss=0.3807, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1493/2855 [1:10:24<45:00,  1.98s/it, loss=0.3807, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1493/2855 [1:10:26<45:00,  1.98s/it, loss=0.1385, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1494/2855 [1:10:26<47:39,  2.10s/it, loss=0.1385, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1494/2855 [1:10:28<47:39,  2.10s/it, loss=0.1857, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1495/2855 [1:10:28<46:04,  2.03s/it, loss=0.1857, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1495/2855 [1:10:30<46:04,  2.03s/it, loss=0.3190, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1496/2855 [1:10:30<43:04,  1.90s/it, loss=0.3190, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1496/2855 [1:10:32<43:04,  1.90s/it, loss=0.2913, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1497/2855 [1:10:32<44:48,  1.98s/it, loss=0.2913, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1497/2855 [1:10:33<44:48,  1.98s/it, loss=0.4982, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1498/2855 [1:10:33<42:18,  1.87s/it, loss=0.4982, lr=2.97e-05]
Epoch 1/8:  52%|█████▏    | 1498/2855 [1:10:35<42:18,  1.87s/it, loss=0.2215, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1499/2855 [1:10:35<41:16,  1.83s/it, loss=0.2215, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1499/2855 [1:10:38<41:16,  1.83s/it, loss=0.3453, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1500/2855 [1:10:38<46:57,  2.08s/it, loss=0.3453, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1500/2855 [1:10:40<46:57,  2.08s/it, loss=0.1365, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1501/2855 [1:10:40<45:59,  2.04s/it, loss=0.1365, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1501/2855 [1:10:42<45:59,  2.04s/it, loss=0.3293, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1502/2855 [1:10:42<48:22,  2.15s/it, loss=0.3293, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1502/2855 [1:10:44<48:22,  2.15s/it, loss=0.0945, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1503/2855 [1:10:44<46:29,  2.06s/it, loss=0.0945, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1503/2855 [1:10:46<46:29,  2.06s/it, loss=0.4626, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1504/2855 [1:10:46<44:48,  1.99s/it, loss=0.4626, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1504/2855 [1:10:48<44:48,  1.99s/it, loss=0.2967, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1505/2855 [1:10:48<46:17,  2.06s/it, loss=0.2967, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1505/2855 [1:10:50<46:17,  2.06s/it, loss=0.1455, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1506/2855 [1:10:50<44:54,  2.00s/it, loss=0.1455, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1506/2855 [1:10:52<44:54,  2.00s/it, loss=0.3338, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1507/2855 [1:10:52<48:30,  2.16s/it, loss=0.3338, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1507/2855 [1:10:54<48:30,  2.16s/it, loss=0.3618, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1508/2855 [1:10:54<46:05,  2.05s/it, loss=0.3618, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1508/2855 [1:10:56<46:05,  2.05s/it, loss=0.2087, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1509/2855 [1:10:56<47:09,  2.10s/it, loss=0.2087, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1509/2855 [1:10:58<47:09,  2.10s/it, loss=0.2154, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1510/2855 [1:10:58<46:47,  2.09s/it, loss=0.2154, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1510/2855 [1:11:00<46:47,  2.09s/it, loss=0.2246, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1511/2855 [1:11:00<44:47,  2.00s/it, loss=0.2246, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1511/2855 [1:11:03<44:47,  2.00s/it, loss=0.1644, lr=2.97e-05]
Epoch 1/8:  53%|██���██▎    | 1512/2855 [1:11:03<46:51,  2.09s/it, loss=0.1644, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1512/2855 [1:11:04<46:51,  2.09s/it, loss=0.1538, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1513/2855 [1:11:04<43:54,  1.96s/it, loss=0.1538, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1513/2855 [1:11:06<43:54,  1.96s/it, loss=0.0712, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1514/2855 [1:11:06<43:34,  1.95s/it, loss=0.0712, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1514/2855 [1:11:08<43:34,  1.95s/it, loss=0.1605, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1515/2855 [1:11:08<44:24,  1.99s/it, loss=0.1605, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1515/2855 [1:11:10<44:24,  1.99s/it, loss=0.1396, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1516/2855 [1:11:10<42:18,  1.90s/it, loss=0.1396, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1516/2855 [1:11:12<42:18,  1.90s/it, loss=0.1962, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1517/2855 [1:11:12<44:43,  2.01s/it, loss=0.1962, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1517/2855 [1:11:14<44:43,  2.01s/it, loss=0.1378, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1518/2855 [1:11:14<41:43,  1.87s/it, loss=0.1378, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1518/2855 [1:11:15<41:43,  1.87s/it, loss=0.2182, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1519/2855 [1:11:15<39:54,  1.79s/it, loss=0.2182, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1519/2855 [1:11:18<39:54,  1.79s/it, loss=0.1333, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1520/2855 [1:11:18<45:20,  2.04s/it, loss=0.1333, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1520/2855 [1:11:20<45:20,  2.04s/it, loss=0.3461, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1521/2855 [1:11:20<45:14,  2.04s/it, loss=0.3461, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1521/2855 [1:11:22<45:14,  2.04s/it, loss=0.3385, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1522/2855 [1:11:22<46:33,  2.10s/it, loss=0.3385, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1522/2855 [1:11:24<46:33,  2.10s/it, loss=0.0611, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1523/2855 [1:11:24<44:39,  2.01s/it, loss=0.0611, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1523/2855 [1:11:26<44:39,  2.01s/it, loss=0.3055, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1524/2855 [1:11:26<43:58,  1.98s/it, loss=0.3055, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1524/2855 [1:11:28<43:58,  1.98s/it, loss=0.0761, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1525/2855 [1:11:28<47:55,  2.16s/it, loss=0.0761, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1525/2855 [1:11:30<47:55,  2.16s/it, loss=0.1901, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1526/2855 [1:11:30<45:48,  2.07s/it, loss=0.1901, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1526/2855 [1:11:33<45:48,  2.07s/it, loss=0.1970, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1527/2855 [1:11:33<46:37,  2.11s/it, loss=0.1970, lr=2.97e-05]
Epoch 1/8:  53%|█████▎    | 1527/2855 [1:11:34<46:37,  2.11s/it, loss=0.1693, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1528/2855 [1:11:34<44:31,  2.01s/it, loss=0.1693, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1528/2855 [1:11:36<44:31,  2.01s/it, loss=0.3949, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1529/2855 [1:11:36<44:23,  2.01s/it, loss=0.3949, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1529/2855 [1:11:39<44:23,  2.01s/it, loss=0.0941, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1530/2855 [1:11:39<46:20,  2.10s/it, loss=0.0941, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1530/2855 [1:11:40<46:20,  2.10s/it, loss=0.3176, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1531/2855 [1:11:40<44:22,  2.01s/it, loss=0.3176, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1531/2855 [1:11:42<44:22,  2.01s/it, loss=0.1920, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1532/2855 [1:11:42<44:35,  2.02s/it, loss=0.1920, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1532/2855 [1:11:44<44:35,  2.02s/it, loss=0.2078, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1533/2855 [1:11:44<40:52,  1.86s/it, loss=0.2078, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1533/2855 [1:11:46<40:52,  1.86s/it, loss=0.0939, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1534/2855 [1:11:46<40:28,  1.84s/it, loss=0.0939, lr=2.97e-05]
Epoch 1/8:  54%|█████▎    | 1534/2855 [1:11:49<40:28,  1.84s/it, loss=0.2762, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1535/2855 [1:11:49<46:34,  2.12s/it, loss=0.2762, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1535/2855 [1:11:50<46:34,  2.12s/it, loss=0.2628, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1536/2855 [1:11:50<44:43,  2.03s/it, loss=0.2628, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1536/2855 [1:11:52<44:43,  2.03s/it, loss=0.2775, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1537/2855 [1:11:52<45:19,  2.06s/it, loss=0.2775, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1537/2855 [1:11:54<45:19,  2.06s/it, loss=0.1126, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1538/2855 [1:11:54<44:13,  2.01s/it, loss=0.1126, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1538/2855 [1:11:56<44:13,  2.01s/it, loss=0.1737, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1539/2855 [1:11:56<43:51,  2.00s/it, loss=0.1737, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1539/2855 [1:11:59<43:51,  2.00s/it, loss=0.1040, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1540/2855 [1:11:59<45:14,  2.06s/it, loss=0.1040, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1540/2855 [1:12:00<45:14,  2.06s/it, loss=0.0743, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1541/2855 [1:12:00<43:05,  1.97s/it, loss=0.0743, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1541/2855 [1:12:03<43:05,  1.97s/it, loss=0.1500, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1542/2855 [1:12:03<44:45,  2.05s/it, loss=0.1500, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1542/2855 [1:12:04<44:45,  2.05s/it, loss=0.2080, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1543/2855 [1:12:04<43:17,  1.98s/it, loss=0.2080, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1543/2855 [1:12:06<43:17,  1.98s/it, loss=0.1627, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1544/2855 [1:12:06<42:20,  1.94s/it, loss=0.1627, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1544/2855 [1:12:09<42:20,  1.94s/it, loss=0.1949, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1545/2855 [1:12:09<44:47,  2.05s/it, loss=0.1949, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1545/2855 [1:12:10<44:47,  2.05s/it, loss=0.0784, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1546/2855 [1:12:10<43:52,  2.01s/it, loss=0.0784, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1546/2855 [1:12:13<43:52,  2.01s/it, loss=0.2966, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1547/2855 [1:12:13<44:30,  2.04s/it, loss=0.2966, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1547/2855 [1:12:14<44:30,  2.04s/it, loss=0.2713, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1548/2855 [1:12:14<43:46,  2.01s/it, loss=0.2713, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1548/2855 [1:12:16<43:46,  2.01s/it, loss=0.2297, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1549/2855 [1:12:16<43:42,  2.01s/it, loss=0.2297, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1549/2855 [1:12:19<43:42,  2.01s/it, loss=0.1981, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1550/2855 [1:12:19<44:56,  2.07s/it, loss=0.1981, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1550/2855 [1:12:20<44:56,  2.07s/it, loss=0.1027, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1551/2855 [1:12:20<42:54,  1.97s/it, loss=0.1027, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1551/2855 [1:12:23<42:54,  1.97s/it, loss=0.4110, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1552/2855 [1:12:23<44:50,  2.07s/it, loss=0.4110, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1552/2855 [1:12:25<44:50,  2.07s/it, loss=0.1425, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1553/2855 [1:12:25<43:22,  2.00s/it, loss=0.1425, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1553/2855 [1:12:26<43:22,  2.00s/it, loss=0.0940, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1554/2855 [1:12:26<42:24,  1.96s/it, loss=0.0940, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1554/2855 [1:12:29<42:24,  1.96s/it, loss=0.3475, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1555/2855 [1:12:29<44:23,  2.05s/it, loss=0.3475, lr=2.97e-05]
Epoch 1/8:  54%|█████▍    | 1555/2855 [1:12:30<44:23,  2.05s/it, loss=0.1287, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1556/2855 [1:12:30<42:23,  1.96s/it, loss=0.1287, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1556/2855 [1:12:33<42:23,  1.96s/it, loss=0.1531, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1557/2855 [1:12:33<44:24,  2.05s/it, loss=0.1531, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1557/2855 [1:12:35<44:24,  2.05s/it, loss=0.1717, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1558/2855 [1:12:35<44:38,  2.06s/it, loss=0.1717, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1558/2855 [1:12:37<44:38,  2.06s/it, loss=0.2604, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1559/2855 [1:12:37<44:14,  2.05s/it, loss=0.2604, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1559/2855 [1:12:39<44:14,  2.05s/it, loss=0.2164, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1560/2855 [1:12:39<44:02,  2.04s/it, loss=0.2164, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1560/2855 [1:12:40<44:02,  2.04s/it, loss=0.3565, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1561/2855 [1:12:40<41:30,  1.92s/it, loss=0.3565, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1561/2855 [1:12:43<41:30,  1.92s/it, loss=0.2606, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1562/2855 [1:12:43<43:58,  2.04s/it, loss=0.2606, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1562/2855 [1:12:45<43:58,  2.04s/it, loss=0.2054, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1563/2855 [1:12:45<43:30,  2.02s/it, loss=0.2054, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1563/2855 [1:12:47<43:30,  2.02s/it, loss=0.1616, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1564/2855 [1:12:47<43:42,  2.03s/it, loss=0.1616, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1564/2855 [1:12:49<43:42,  2.03s/it, loss=0.0336, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1565/2855 [1:12:49<44:16,  2.06s/it, loss=0.0336, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1565/2855 [1:12:51<44:16,  2.06s/it, loss=0.1371, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1566/2855 [1:12:51<42:32,  1.98s/it, loss=0.1371, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1566/2855 [1:12:53<42:32,  1.98s/it, loss=0.1232, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1567/2855 [1:12:53<43:21,  2.02s/it, loss=0.1232, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1567/2855 [1:12:55<43:21,  2.02s/it, loss=0.1620, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1568/2855 [1:12:55<42:17,  1.97s/it, loss=0.1620, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1568/2855 [1:12:56<42:17,  1.97s/it, loss=0.0690, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1569/2855 [1:12:56<40:58,  1.91s/it, loss=0.0690, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1569/2855 [1:12:59<40:58,  1.91s/it, loss=0.2080, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1570/2855 [1:12:59<43:29,  2.03s/it, loss=0.2080, lr=2.97e-05]
Epoch 1/8:  55%|█████▍    | 1570/2855 [1:13:01<43:29,  2.03s/it, loss=0.2861, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1571/2855 [1:13:01<42:12,  1.97s/it, loss=0.2861, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1571/2855 [1:13:03<42:12,  1.97s/it, loss=0.0786, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1572/2855 [1:13:03<43:23,  2.03s/it, loss=0.0786, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1572/2855 [1:13:05<43:23,  2.03s/it, loss=0.3174, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1573/2855 [1:13:05<42:19,  1.98s/it, loss=0.3174, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1573/2855 [1:13:06<42:19,  1.98s/it, loss=0.2419, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1574/2855 [1:13:06<40:46,  1.91s/it, loss=0.2419, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1574/2855 [1:13:09<40:46,  1.91s/it, loss=0.3937, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1575/2855 [1:13:09<42:47,  2.01s/it, loss=0.3937, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1575/2855 [1:13:10<42:47,  2.01s/it, loss=0.1850, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1576/2855 [1:13:10<40:21,  1.89s/it, loss=0.1850, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1576/2855 [1:13:12<40:21,  1.89s/it, loss=0.2090, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1577/2855 [1:13:12<40:08,  1.88s/it, loss=0.2090, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1577/2855 [1:13:14<40:08,  1.88s/it, loss=0.1782, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1578/2855 [1:13:14<39:52,  1.87s/it, loss=0.1782, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1578/2855 [1:13:15<39:52,  1.87s/it, loss=0.4226, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1579/2855 [1:13:15<37:24,  1.76s/it, loss=0.4226, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1579/2855 [1:13:17<37:24,  1.76s/it, loss=0.3062, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1580/2855 [1:13:17<38:20,  1.80s/it, loss=0.3062, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1580/2855 [1:13:20<38:20,  1.80s/it, loss=0.2452, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1581/2855 [1:13:20<40:47,  1.92s/it, loss=0.2452, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1581/2855 [1:13:21<40:47,  1.92s/it, loss=0.1119, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1582/2855 [1:13:21<40:22,  1.90s/it, loss=0.1119, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1582/2855 [1:13:24<40:22,  1.90s/it, loss=0.2106, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1583/2855 [1:13:24<42:02,  1.98s/it, loss=0.2106, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1583/2855 [1:13:25<42:02,  1.98s/it, loss=0.3847, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1584/2855 [1:13:25<40:26,  1.91s/it, loss=0.3847, lr=2.97e-05]
Epoch 1/8:  55%|█████▌    | 1584/2855 [1:13:27<40:26,  1.91s/it, loss=0.0338, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1585/2855 [1:13:27<41:14,  1.95s/it, loss=0.0338, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1585/2855 [1:13:30<41:14,  1.95s/it, loss=0.2506, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1586/2855 [1:13:30<42:15,  2.00s/it, loss=0.2506, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1586/2855 [1:13:31<42:15,  2.00s/it, loss=0.3566, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1587/2855 [1:13:31<40:06,  1.90s/it, loss=0.3566, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1587/2855 [1:13:33<40:06,  1.90s/it, loss=0.1452, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1588/2855 [1:13:33<42:17,  2.00s/it, loss=0.1452, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1588/2855 [1:13:35<42:17,  2.00s/it, loss=0.1584, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1589/2855 [1:13:35<41:10,  1.95s/it, loss=0.1584, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1589/2855 [1:13:37<41:10,  1.95s/it, loss=0.3200, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1590/2855 [1:13:37<40:36,  1.93s/it, loss=0.3200, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1590/2855 [1:13:39<40:36,  1.93s/it, loss=0.2026, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1591/2855 [1:13:39<42:30,  2.02s/it, loss=0.2026, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1591/2855 [1:13:41<42:30,  2.02s/it, loss=0.2240, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1592/2855 [1:13:41<40:43,  1.93s/it, loss=0.2240, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1592/2855 [1:13:43<40:43,  1.93s/it, loss=0.1847, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1593/2855 [1:13:43<41:26,  1.97s/it, loss=0.1847, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1593/2855 [1:13:45<41:26,  1.97s/it, loss=0.2916, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1594/2855 [1:13:45<39:44,  1.89s/it, loss=0.2916, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1594/2855 [1:13:46<39:44,  1.89s/it, loss=0.2234, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1595/2855 [1:13:46<37:54,  1.81s/it, loss=0.2234, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1595/2855 [1:13:49<37:54,  1.81s/it, loss=0.2350, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1596/2855 [1:13:49<40:11,  1.92s/it, loss=0.2350, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1596/2855 [1:13:50<40:11,  1.92s/it, loss=0.1515, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1597/2855 [1:13:50<38:02,  1.81s/it, loss=0.1515, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1597/2855 [1:13:52<38:02,  1.81s/it, loss=0.2142, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1598/2855 [1:13:52<37:29,  1.79s/it, loss=0.2142, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1598/2855 [1:13:54<37:29,  1.79s/it, loss=0.3710, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1599/2855 [1:13:54<42:14,  2.02s/it, loss=0.3710, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1599/2855 [1:13:56<42:14,  2.02s/it, loss=0.2278, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1600/2855 [1:13:56<40:55,  1.96s/it, loss=0.2278, lr=2.97e-05]
Epoch 1/8:  56%|█████▌    | 1600/2855 [1:13:58<40:55,  1.96s/it, loss=0.2791, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1601/2855 [1:13:58<42:14,  2.02s/it, loss=0.2791, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1601/2855 [1:14:00<42:14,  2.02s/it, loss=0.1530, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1602/2855 [1:14:00<41:19,  1.98s/it, loss=0.1530, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1602/2855 [1:14:02<41:19,  1.98s/it, loss=0.1871, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1603/2855 [1:14:02<39:39,  1.90s/it, loss=0.1871, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1603/2855 [1:14:04<39:39,  1.90s/it, loss=0.2270, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1604/2855 [1:14:04<41:52,  2.01s/it, loss=0.2270, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1604/2855 [1:14:06<41:52,  2.01s/it, loss=0.1463, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1605/2855 [1:14:06<40:22,  1.94s/it, loss=0.1463, lr=2.96e-05]
Epoch 1/8:  56%|█████▌    | 1605/2855 [1:14:08<40:22,  1.94s/it, loss=0.0993, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1606/2855 [1:14:08<41:18,  1.98s/it, loss=0.0993, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1606/2855 [1:14:10<41:18,  1.98s/it, loss=0.1383, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1607/2855 [1:14:10<41:10,  1.98s/it, loss=0.1383, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1607/2855 [1:14:12<41:10,  1.98s/it, loss=0.2808, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1608/2855 [1:14:12<39:53,  1.92s/it, loss=0.2808, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1608/2855 [1:14:14<39:53,  1.92s/it, loss=0.3299, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1609/2855 [1:14:14<41:22,  1.99s/it, loss=0.3299, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1609/2855 [1:14:16<41:22,  1.99s/it, loss=0.1818, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1610/2855 [1:14:16<38:23,  1.85s/it, loss=0.1818, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1610/2855 [1:14:17<38:23,  1.85s/it, loss=0.2350, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1611/2855 [1:14:17<36:28,  1.76s/it, loss=0.2350, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1611/2855 [1:14:19<36:28,  1.76s/it, loss=0.2976, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1612/2855 [1:14:19<39:06,  1.89s/it, loss=0.2976, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1612/2855 [1:14:21<39:06,  1.89s/it, loss=0.2070, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1613/2855 [1:14:21<39:13,  1.89s/it, loss=0.2070, lr=2.96e-05]
Epoch 1/8:  56%|█████▋    | 1613/2855 [1:14:24<39:13,  1.89s/it, loss=0.1029, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1614/2855 [1:14:24<41:37,  2.01s/it, loss=0.1029, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1614/2855 [1:14:25<41:37,  2.01s/it, loss=0.1511, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1615/2855 [1:14:25<40:30,  1.96s/it, loss=0.1511, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1615/2855 [1:14:27<40:30,  1.96s/it, loss=0.1762, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1616/2855 [1:14:27<38:57,  1.89s/it, loss=0.1762, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1616/2855 [1:14:30<38:57,  1.89s/it, loss=0.3861, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1617/2855 [1:14:30<42:12,  2.05s/it, loss=0.3861, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1617/2855 [1:14:31<42:12,  2.05s/it, loss=0.1826, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1618/2855 [1:14:31<40:49,  1.98s/it, loss=0.1826, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1618/2855 [1:14:33<40:49,  1.98s/it, loss=0.1565, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1619/2855 [1:14:33<40:42,  1.98s/it, loss=0.1565, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1619/2855 [1:14:35<40:42,  1.98s/it, loss=0.2717, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1620/2855 [1:14:35<40:22,  1.96s/it, loss=0.2717, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1620/2855 [1:14:37<40:22,  1.96s/it, loss=0.1676, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1621/2855 [1:14:37<39:24,  1.92s/it, loss=0.1676, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1621/2855 [1:14:39<39:24,  1.92s/it, loss=0.1214, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1622/2855 [1:14:39<41:34,  2.02s/it, loss=0.1214, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1622/2855 [1:14:41<41:34,  2.02s/it, loss=0.1491, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1623/2855 [1:14:41<41:27,  2.02s/it, loss=0.1491, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1623/2855 [1:14:43<41:27,  2.02s/it, loss=0.2953, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1624/2855 [1:14:44<42:16,  2.06s/it, loss=0.2953, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1624/2855 [1:14:45<42:16,  2.06s/it, loss=0.0547, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1625/2855 [1:14:45<40:07,  1.96s/it, loss=0.0547, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1625/2855 [1:14:47<40:07,  1.96s/it, loss=0.1618, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1626/2855 [1:14:47<37:43,  1.84s/it, loss=0.1618, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1626/2855 [1:14:49<37:43,  1.84s/it, loss=0.1086, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1627/2855 [1:14:49<39:45,  1.94s/it, loss=0.1086, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1627/2855 [1:14:51<39:45,  1.94s/it, loss=0.1370, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1628/2855 [1:14:51<37:59,  1.86s/it, loss=0.1370, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1628/2855 [1:14:52<37:59,  1.86s/it, loss=0.1999, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1629/2855 [1:14:52<36:20,  1.78s/it, loss=0.1999, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1629/2855 [1:14:54<36:20,  1.78s/it, loss=0.0487, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1630/2855 [1:14:54<38:49,  1.90s/it, loss=0.0487, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1630/2855 [1:14:56<38:49,  1.90s/it, loss=0.1117, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1631/2855 [1:14:56<36:32,  1.79s/it, loss=0.1117, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1631/2855 [1:14:58<36:32,  1.79s/it, loss=0.2269, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1632/2855 [1:14:58<35:08,  1.72s/it, loss=0.2269, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1632/2855 [1:15:00<35:08,  1.72s/it, loss=0.1676, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1633/2855 [1:15:00<37:54,  1.86s/it, loss=0.1676, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1633/2855 [1:15:01<37:54,  1.86s/it, loss=0.1497, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1634/2855 [1:15:01<35:43,  1.76s/it, loss=0.1497, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1634/2855 [1:15:03<35:43,  1.76s/it, loss=0.3875, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1635/2855 [1:15:03<34:54,  1.72s/it, loss=0.3875, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1635/2855 [1:15:05<34:54,  1.72s/it, loss=0.1692, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1636/2855 [1:15:05<37:34,  1.85s/it, loss=0.1692, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1636/2855 [1:15:06<37:34,  1.85s/it, loss=0.1693, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1637/2855 [1:15:06<35:23,  1.74s/it, loss=0.1693, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1637/2855 [1:15:08<35:23,  1.74s/it, loss=0.0628, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1638/2855 [1:15:08<35:09,  1.73s/it, loss=0.0628, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1638/2855 [1:15:11<35:09,  1.73s/it, loss=0.2881, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1639/2855 [1:15:11<39:13,  1.94s/it, loss=0.2881, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1639/2855 [1:15:12<39:13,  1.94s/it, loss=0.2809, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1640/2855 [1:15:12<38:52,  1.92s/it, loss=0.2809, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1640/2855 [1:15:15<38:52,  1.92s/it, loss=0.2497, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1641/2855 [1:15:15<41:01,  2.03s/it, loss=0.2497, lr=2.96e-05]
Epoch 1/8:  57%|█████▋    | 1641/2855 [1:15:17<41:01,  2.03s/it, loss=0.1316, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1642/2855 [1:15:17<39:32,  1.96s/it, loss=0.1316, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1642/2855 [1:15:18<39:32,  1.96s/it, loss=0.2645, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1643/2855 [1:15:18<39:10,  1.94s/it, loss=0.2645, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1643/2855 [1:15:21<39:10,  1.94s/it, loss=0.3236, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1644/2855 [1:15:21<40:26,  2.00s/it, loss=0.3236, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1644/2855 [1:15:22<40:26,  2.00s/it, loss=0.4688, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1645/2855 [1:15:22<38:52,  1.93s/it, loss=0.4688, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1645/2855 [1:15:25<38:52,  1.93s/it, loss=0.2055, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1646/2855 [1:15:25<40:17,  2.00s/it, loss=0.2055, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1646/2855 [1:15:26<40:17,  2.00s/it, loss=0.0997, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1647/2855 [1:15:26<39:19,  1.95s/it, loss=0.0997, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1647/2855 [1:15:28<39:19,  1.95s/it, loss=0.2934, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1648/2855 [1:15:28<38:21,  1.91s/it, loss=0.2934, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1648/2855 [1:15:30<38:21,  1.91s/it, loss=0.2145, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1649/2855 [1:15:30<40:44,  2.03s/it, loss=0.2145, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1649/2855 [1:15:32<40:44,  2.03s/it, loss=0.3067, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1650/2855 [1:15:32<39:18,  1.96s/it, loss=0.3067, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1650/2855 [1:15:35<39:18,  1.96s/it, loss=0.2549, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1651/2855 [1:15:35<41:09,  2.05s/it, loss=0.2549, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1651/2855 [1:15:36<41:09,  2.05s/it, loss=0.3754, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1652/2855 [1:15:36<39:39,  1.98s/it, loss=0.3754, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1652/2855 [1:15:38<39:39,  1.98s/it, loss=0.1587, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1653/2855 [1:15:38<37:22,  1.87s/it, loss=0.1587, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1653/2855 [1:15:40<37:22,  1.87s/it, loss=0.1834, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1654/2855 [1:15:40<38:46,  1.94s/it, loss=0.1834, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1654/2855 [1:15:42<38:46,  1.94s/it, loss=0.2313, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1655/2855 [1:15:42<37:20,  1.87s/it, loss=0.2313, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1655/2855 [1:15:44<37:20,  1.87s/it, loss=0.1593, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1656/2855 [1:15:44<37:14,  1.86s/it, loss=0.1593, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1656/2855 [1:15:46<37:14,  1.86s/it, loss=0.1731, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1657/2855 [1:15:46<38:19,  1.92s/it, loss=0.1731, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1657/2855 [1:15:48<38:19,  1.92s/it, loss=0.0098, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1658/2855 [1:15:48<39:59,  2.00s/it, loss=0.0098, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1658/2855 [1:15:50<39:59,  2.00s/it, loss=0.3205, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1659/2855 [1:15:50<42:12,  2.12s/it, loss=0.3205, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1659/2855 [1:15:52<42:12,  2.12s/it, loss=0.1722, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1660/2855 [1:15:52<40:04,  2.01s/it, loss=0.1722, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1660/2855 [1:15:54<40:04,  2.01s/it, loss=0.2012, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1661/2855 [1:15:54<40:04,  2.01s/it, loss=0.2012, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1661/2855 [1:15:57<40:04,  2.01s/it, loss=0.2132, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1662/2855 [1:15:57<43:55,  2.21s/it, loss=0.2132, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1662/2855 [1:15:59<43:55,  2.21s/it, loss=0.0773, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1663/2855 [1:15:59<44:28,  2.24s/it, loss=0.0773, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1663/2855 [1:16:01<44:28,  2.24s/it, loss=0.1534, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1664/2855 [1:16:01<42:45,  2.15s/it, loss=0.1534, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1664/2855 [1:16:03<42:45,  2.15s/it, loss=0.2473, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1665/2855 [1:16:03<40:43,  2.05s/it, loss=0.2473, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1665/2855 [1:16:05<40:43,  2.05s/it, loss=0.0407, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1666/2855 [1:16:05<41:20,  2.09s/it, loss=0.0407, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1666/2855 [1:16:07<41:20,  2.09s/it, loss=0.2570, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1667/2855 [1:16:07<40:03,  2.02s/it, loss=0.2570, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1667/2855 [1:16:09<40:03,  2.02s/it, loss=0.0268, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1668/2855 [1:16:09<39:13,  1.98s/it, loss=0.0268, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1668/2855 [1:16:11<39:13,  1.98s/it, loss=0.0881, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1669/2855 [1:16:11<40:11,  2.03s/it, loss=0.0881, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1669/2855 [1:16:12<40:11,  2.03s/it, loss=0.1919, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1670/2855 [1:16:12<37:10,  1.88s/it, loss=0.1919, lr=2.96e-05]
Epoch 1/8:  58%|█████▊    | 1670/2855 [1:16:14<37:10,  1.88s/it, loss=0.1352, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1671/2855 [1:16:14<37:37,  1.91s/it, loss=0.1352, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1671/2855 [1:16:17<37:37,  1.91s/it, loss=0.1536, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1672/2855 [1:16:17<39:31,  2.00s/it, loss=0.1536, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1672/2855 [1:16:18<39:31,  2.00s/it, loss=0.0700, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1673/2855 [1:16:18<38:27,  1.95s/it, loss=0.0700, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1673/2855 [1:16:21<38:27,  1.95s/it, loss=0.3078, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1674/2855 [1:16:21<40:54,  2.08s/it, loss=0.3078, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1674/2855 [1:16:23<40:54,  2.08s/it, loss=0.2812, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1675/2855 [1:16:23<39:13,  1.99s/it, loss=0.2812, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1675/2855 [1:16:25<39:13,  1.99s/it, loss=0.0405, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1676/2855 [1:16:25<40:10,  2.04s/it, loss=0.0405, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1676/2855 [1:16:27<40:10,  2.04s/it, loss=0.3182, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1677/2855 [1:16:27<39:17,  2.00s/it, loss=0.3182, lr=2.96e-05]
Epoch 1/8:  59%|█████▊    | 1677/2855 [1:16:28<39:17,  2.00s/it, loss=0.3540, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1678/2855 [1:16:28<37:48,  1.93s/it, loss=0.3540, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1678/2855 [1:16:31<37:48,  1.93s/it, loss=0.2519, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1679/2855 [1:16:31<39:53,  2.03s/it, loss=0.2519, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1679/2855 [1:16:32<39:53,  2.03s/it, loss=0.2275, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1680/2855 [1:16:32<38:19,  1.96s/it, loss=0.2275, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1680/2855 [1:16:35<38:19,  1.96s/it, loss=0.2555, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1681/2855 [1:16:35<38:54,  1.99s/it, loss=0.2555, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1681/2855 [1:16:37<38:54,  1.99s/it, loss=0.2506, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1682/2855 [1:16:37<39:41,  2.03s/it, loss=0.2506, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1682/2855 [1:16:38<39:41,  2.03s/it, loss=0.4010, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1683/2855 [1:16:38<38:00,  1.95s/it, loss=0.4010, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1683/2855 [1:16:40<38:00,  1.95s/it, loss=0.1810, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1684/2855 [1:16:40<38:35,  1.98s/it, loss=0.1810, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1684/2855 [1:16:42<38:35,  1.98s/it, loss=0.2144, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1685/2855 [1:16:42<36:32,  1.87s/it, loss=0.2144, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1685/2855 [1:16:44<36:32,  1.87s/it, loss=0.1881, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1686/2855 [1:16:44<34:51,  1.79s/it, loss=0.1881, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1686/2855 [1:16:46<34:51,  1.79s/it, loss=0.0829, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1687/2855 [1:16:46<40:01,  2.06s/it, loss=0.0829, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1687/2855 [1:16:48<40:01,  2.06s/it, loss=0.1415, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1688/2855 [1:16:48<38:55,  2.00s/it, loss=0.1415, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1688/2855 [1:16:50<38:55,  2.00s/it, loss=0.1942, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1689/2855 [1:16:50<39:57,  2.06s/it, loss=0.1942, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1689/2855 [1:16:53<39:57,  2.06s/it, loss=0.2097, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1690/2855 [1:16:53<40:30,  2.09s/it, loss=0.2097, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1690/2855 [1:16:55<40:30,  2.09s/it, loss=0.1749, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1691/2855 [1:16:55<40:07,  2.07s/it, loss=0.1749, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1691/2855 [1:16:57<40:07,  2.07s/it, loss=0.1344, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1692/2855 [1:16:57<39:22,  2.03s/it, loss=0.1344, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1692/2855 [1:16:58<39:22,  2.03s/it, loss=0.1766, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1693/2855 [1:16:58<37:49,  1.95s/it, loss=0.1766, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1693/2855 [1:17:01<37:49,  1.95s/it, loss=0.3853, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1694/2855 [1:17:01<39:48,  2.06s/it, loss=0.3853, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1694/2855 [1:17:03<39:48,  2.06s/it, loss=0.1315, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1695/2855 [1:17:03<38:52,  2.01s/it, loss=0.1315, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1695/2855 [1:17:05<38:52,  2.01s/it, loss=0.1942, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1696/2855 [1:17:05<40:20,  2.09s/it, loss=0.1942, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1696/2855 [1:17:07<40:20,  2.09s/it, loss=0.2227, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1697/2855 [1:17:07<39:35,  2.05s/it, loss=0.2227, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1697/2855 [1:17:09<39:35,  2.05s/it, loss=0.1034, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1698/2855 [1:17:09<37:56,  1.97s/it, loss=0.1034, lr=2.96e-05]
Epoch 1/8:  59%|█████▉    | 1698/2855 [1:17:11<37:56,  1.97s/it, loss=0.2168, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1699/2855 [1:17:11<39:43,  2.06s/it, loss=0.2168, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1699/2855 [1:17:13<39:43,  2.06s/it, loss=0.1623, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1700/2855 [1:17:13<37:54,  1.97s/it, loss=0.1623, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1700/2855 [1:17:14<37:54,  1.97s/it, loss=0.1886, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1701/2855 [1:17:14<36:28,  1.90s/it, loss=0.1886, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1701/2855 [1:17:16<36:28,  1.90s/it, loss=0.2115, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1702/2855 [1:17:16<36:53,  1.92s/it, loss=0.2115, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1702/2855 [1:17:18<36:53,  1.92s/it, loss=0.2257, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1703/2855 [1:17:18<34:30,  1.80s/it, loss=0.2257, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1703/2855 [1:17:20<34:30,  1.80s/it, loss=0.1271, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1704/2855 [1:17:20<37:15,  1.94s/it, loss=0.1271, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1704/2855 [1:17:22<37:15,  1.94s/it, loss=0.2334, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1705/2855 [1:17:22<39:10,  2.04s/it, loss=0.2334, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1705/2855 [1:17:24<39:10,  2.04s/it, loss=0.2431, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1706/2855 [1:17:24<38:00,  1.99s/it, loss=0.2431, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1706/2855 [1:17:26<38:00,  1.99s/it, loss=0.1322, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1707/2855 [1:17:26<39:02,  2.04s/it, loss=0.1322, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1707/2855 [1:17:28<39:02,  2.04s/it, loss=0.1850, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1708/2855 [1:17:28<37:40,  1.97s/it, loss=0.1850, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1708/2855 [1:17:30<37:40,  1.97s/it, loss=0.1581, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1709/2855 [1:17:30<38:42,  2.03s/it, loss=0.1581, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1709/2855 [1:17:32<38:42,  2.03s/it, loss=0.2401, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1710/2855 [1:17:32<38:55,  2.04s/it, loss=0.2401, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1710/2855 [1:17:34<38:55,  2.04s/it, loss=0.2003, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1711/2855 [1:17:34<37:27,  1.96s/it, loss=0.2003, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1711/2855 [1:17:37<37:27,  1.96s/it, loss=0.2479, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1712/2855 [1:17:37<39:31,  2.08s/it, loss=0.2479, lr=2.96e-05]
Epoch 1/8:  60%|█████▉    | 1712/2855 [1:17:38<39:31,  2.08s/it, loss=0.1209, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1713/2855 [1:17:38<38:27,  2.02s/it, loss=0.1209, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1713/2855 [1:17:41<38:27,  2.02s/it, loss=0.1146, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1714/2855 [1:17:41<39:02,  2.05s/it, loss=0.1146, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1714/2855 [1:17:43<39:02,  2.05s/it, loss=0.2255, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1715/2855 [1:17:43<38:43,  2.04s/it, loss=0.2255, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1715/2855 [1:17:44<38:43,  2.04s/it, loss=0.1006, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1716/2855 [1:17:44<36:54,  1.94s/it, loss=0.1006, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1716/2855 [1:17:46<36:54,  1.94s/it, loss=0.2339, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1717/2855 [1:17:46<37:46,  1.99s/it, loss=0.2339, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1717/2855 [1:17:48<37:46,  1.99s/it, loss=0.0931, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1718/2855 [1:17:48<35:37,  1.88s/it, loss=0.0931, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1718/2855 [1:17:51<35:37,  1.88s/it, loss=0.1323, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1719/2855 [1:17:51<40:43,  2.15s/it, loss=0.1323, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1719/2855 [1:17:53<40:43,  2.15s/it, loss=0.0339, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1720/2855 [1:17:53<39:53,  2.11s/it, loss=0.0339, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1720/2855 [1:17:55<39:53,  2.11s/it, loss=0.1089, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1721/2855 [1:17:55<40:13,  2.13s/it, loss=0.1089, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1721/2855 [1:17:57<40:13,  2.13s/it, loss=0.1328, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1722/2855 [1:17:57<41:03,  2.17s/it, loss=0.1328, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1722/2855 [1:17:59<41:03,  2.17s/it, loss=0.0440, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1723/2855 [1:17:59<39:06,  2.07s/it, loss=0.0440, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1723/2855 [1:18:01<39:06,  2.07s/it, loss=0.0892, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1724/2855 [1:18:01<40:08,  2.13s/it, loss=0.0892, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1724/2855 [1:18:03<40:08,  2.13s/it, loss=0.1504, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1725/2855 [1:18:03<39:04,  2.07s/it, loss=0.1504, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1725/2855 [1:18:05<39:04,  2.07s/it, loss=0.1540, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1726/2855 [1:18:05<38:59,  2.07s/it, loss=0.1540, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1726/2855 [1:18:08<38:59,  2.07s/it, loss=0.1621, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1727/2855 [1:18:08<39:44,  2.11s/it, loss=0.1621, lr=2.96e-05]
Epoch 1/8:  60%|██████    | 1727/2855 [1:18:09<39:44,  2.11s/it, loss=0.2572, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1728/2855 [1:18:09<38:28,  2.05s/it, loss=0.2572, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1728/2855 [1:18:12<38:28,  2.05s/it, loss=0.2208, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1729/2855 [1:18:12<40:42,  2.17s/it, loss=0.2208, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1729/2855 [1:18:14<40:42,  2.17s/it, loss=0.0962, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1730/2855 [1:18:14<38:50,  2.07s/it, loss=0.0962, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1730/2855 [1:18:16<38:50,  2.07s/it, loss=0.2540, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1731/2855 [1:18:16<39:19,  2.10s/it, loss=0.2540, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1731/2855 [1:18:18<39:19,  2.10s/it, loss=0.2788, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1732/2855 [1:18:18<39:15,  2.10s/it, loss=0.2788, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1732/2855 [1:18:20<39:15,  2.10s/it, loss=0.2749, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1733/2855 [1:18:20<39:03,  2.09s/it, loss=0.2749, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1733/2855 [1:18:22<39:03,  2.09s/it, loss=0.2883, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1734/2855 [1:18:22<39:29,  2.11s/it, loss=0.2883, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1734/2855 [1:18:24<39:29,  2.11s/it, loss=0.1511, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1735/2855 [1:18:24<37:52,  2.03s/it, loss=0.1511, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1735/2855 [1:18:27<37:52,  2.03s/it, loss=0.2017, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1736/2855 [1:18:27<40:15,  2.16s/it, loss=0.2017, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1736/2855 [1:18:29<40:15,  2.16s/it, loss=0.1477, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1737/2855 [1:18:29<39:21,  2.11s/it, loss=0.1477, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1737/2855 [1:18:31<39:21,  2.11s/it, loss=0.2493, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1738/2855 [1:18:31<39:15,  2.11s/it, loss=0.2493, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1738/2855 [1:18:33<39:15,  2.11s/it, loss=0.1958, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1739/2855 [1:18:33<39:00,  2.10s/it, loss=0.1958, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1739/2855 [1:18:35<39:00,  2.10s/it, loss=0.1493, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1740/2855 [1:18:35<37:24,  2.01s/it, loss=0.1493, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1740/2855 [1:18:37<37:24,  2.01s/it, loss=0.1924, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1741/2855 [1:18:37<40:42,  2.19s/it, loss=0.1924, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1741/2855 [1:18:39<40:42,  2.19s/it, loss=0.1549, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1742/2855 [1:18:39<38:56,  2.10s/it, loss=0.1549, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1742/2855 [1:18:41<38:56,  2.10s/it, loss=0.1272, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1743/2855 [1:18:41<38:36,  2.08s/it, loss=0.1272, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1743/2855 [1:18:43<38:36,  2.08s/it, loss=0.1973, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1744/2855 [1:18:43<38:02,  2.05s/it, loss=0.1973, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1744/2855 [1:18:45<38:02,  2.05s/it, loss=0.2520, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1745/2855 [1:18:45<35:08,  1.90s/it, loss=0.2520, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1745/2855 [1:18:47<35:08,  1.90s/it, loss=0.3261, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1746/2855 [1:18:47<37:32,  2.03s/it, loss=0.3261, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1746/2855 [1:18:49<37:32,  2.03s/it, loss=0.1020, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1747/2855 [1:18:49<35:54,  1.94s/it, loss=0.1020, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1747/2855 [1:18:51<35:54,  1.94s/it, loss=0.3324, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1748/2855 [1:18:51<35:43,  1.94s/it, loss=0.3324, lr=2.96e-05]
Epoch 1/8:  61%|██████    | 1748/2855 [1:18:53<35:43,  1.94s/it, loss=0.1041, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1749/2855 [1:18:53<36:26,  1.98s/it, loss=0.1041, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1749/2855 [1:18:54<36:26,  1.98s/it, loss=0.1443, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1750/2855 [1:18:54<34:25,  1.87s/it, loss=0.1443, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1750/2855 [1:18:57<34:25,  1.87s/it, loss=0.1613, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1751/2855 [1:18:57<40:09,  2.18s/it, loss=0.1613, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1751/2855 [1:18:59<40:09,  2.18s/it, loss=0.2395, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1752/2855 [1:18:59<40:34,  2.21s/it, loss=0.2395, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1752/2855 [1:19:02<40:34,  2.21s/it, loss=0.1965, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1753/2855 [1:19:02<41:16,  2.25s/it, loss=0.1965, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1753/2855 [1:19:04<41:16,  2.25s/it, loss=0.0629, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1754/2855 [1:19:04<39:33,  2.16s/it, loss=0.0629, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1754/2855 [1:19:06<39:33,  2.16s/it, loss=0.3350, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1755/2855 [1:19:06<38:57,  2.12s/it, loss=0.3350, lr=2.96e-05]
Epoch 1/8:  61%|██████▏   | 1755/2855 [1:19:08<38:57,  2.12s/it, loss=0.0881, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1756/2855 [1:19:08<38:56,  2.13s/it, loss=0.0881, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1756/2855 [1:19:10<38:56,  2.13s/it, loss=0.0348, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1757/2855 [1:19:10<37:55,  2.07s/it, loss=0.0348, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1757/2855 [1:19:12<37:55,  2.07s/it, loss=0.1479, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1758/2855 [1:19:12<39:53,  2.18s/it, loss=0.1479, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1758/2855 [1:19:14<39:53,  2.18s/it, loss=0.1700, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1759/2855 [1:19:14<38:21,  2.10s/it, loss=0.1700, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1759/2855 [1:19:16<38:21,  2.10s/it, loss=0.1813, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1760/2855 [1:19:16<38:51,  2.13s/it, loss=0.1813, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1760/2855 [1:19:19<38:51,  2.13s/it, loss=0.2935, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1761/2855 [1:19:19<38:40,  2.12s/it, loss=0.2935, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1761/2855 [1:19:21<38:40,  2.12s/it, loss=0.2958, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1762/2855 [1:19:21<38:03,  2.09s/it, loss=0.2958, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1762/2855 [1:19:23<38:03,  2.09s/it, loss=0.1342, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1763/2855 [1:19:23<38:41,  2.13s/it, loss=0.1342, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1763/2855 [1:19:24<38:41,  2.13s/it, loss=0.0531, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1764/2855 [1:19:24<35:59,  1.98s/it, loss=0.0531, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1764/2855 [1:19:26<35:59,  1.98s/it, loss=0.2485, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1765/2855 [1:19:26<36:07,  1.99s/it, loss=0.2485, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1765/2855 [1:19:28<36:07,  1.99s/it, loss=0.0496, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1766/2855 [1:19:28<36:01,  1.98s/it, loss=0.0496, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1766/2855 [1:19:30<36:01,  1.98s/it, loss=0.3770, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1767/2855 [1:19:30<33:56,  1.87s/it, loss=0.3770, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1767/2855 [1:19:33<33:56,  1.87s/it, loss=0.3409, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1768/2855 [1:19:33<38:33,  2.13s/it, loss=0.3409, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1768/2855 [1:19:35<38:33,  2.13s/it, loss=0.1488, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1769/2855 [1:19:35<37:55,  2.10s/it, loss=0.1488, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1769/2855 [1:19:37<37:55,  2.10s/it, loss=0.1839, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1770/2855 [1:19:37<38:11,  2.11s/it, loss=0.1839, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1770/2855 [1:19:39<38:11,  2.11s/it, loss=0.2213, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1771/2855 [1:19:39<37:27,  2.07s/it, loss=0.2213, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1771/2855 [1:19:41<37:27,  2.07s/it, loss=0.1061, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1772/2855 [1:19:41<37:20,  2.07s/it, loss=0.1061, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1772/2855 [1:19:43<37:20,  2.07s/it, loss=0.1834, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1773/2855 [1:19:43<38:08,  2.11s/it, loss=0.1834, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1773/2855 [1:19:45<38:08,  2.11s/it, loss=0.2522, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1774/2855 [1:19:45<36:33,  2.03s/it, loss=0.2522, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1774/2855 [1:19:47<36:33,  2.03s/it, loss=0.2390, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1775/2855 [1:19:47<38:25,  2.14s/it, loss=0.2390, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1775/2855 [1:19:49<38:25,  2.14s/it, loss=0.2254, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1776/2855 [1:19:49<37:42,  2.10s/it, loss=0.2254, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1776/2855 [1:19:52<37:42,  2.10s/it, loss=0.2077, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1777/2855 [1:19:52<38:02,  2.12s/it, loss=0.2077, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1777/2855 [1:19:54<38:02,  2.12s/it, loss=0.1402, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1778/2855 [1:19:54<38:12,  2.13s/it, loss=0.1402, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1778/2855 [1:19:56<38:12,  2.13s/it, loss=0.3132, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1779/2855 [1:19:56<37:12,  2.07s/it, loss=0.3132, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1779/2855 [1:19:58<37:12,  2.07s/it, loss=0.1271, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1780/2855 [1:19:58<37:35,  2.10s/it, loss=0.1271, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1780/2855 [1:20:00<37:35,  2.10s/it, loss=0.0903, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1781/2855 [1:20:00<35:47,  2.00s/it, loss=0.0903, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1781/2855 [1:20:02<35:47,  2.00s/it, loss=0.3140, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1782/2855 [1:20:02<35:44,  2.00s/it, loss=0.3140, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1782/2855 [1:20:04<35:44,  2.00s/it, loss=0.1517, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1783/2855 [1:20:04<38:43,  2.17s/it, loss=0.1517, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1783/2855 [1:20:06<38:43,  2.17s/it, loss=0.1718, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1784/2855 [1:20:06<36:36,  2.05s/it, loss=0.1718, lr=2.96e-05]
Epoch 1/8:  62%|██████▏   | 1784/2855 [1:20:08<36:36,  2.05s/it, loss=0.1897, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1785/2855 [1:20:08<36:29,  2.05s/it, loss=0.1897, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1785/2855 [1:20:10<36:29,  2.05s/it, loss=0.0226, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1786/2855 [1:20:10<36:04,  2.02s/it, loss=0.0226, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1786/2855 [1:20:13<36:04,  2.02s/it, loss=0.2375, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1787/2855 [1:20:13<44:04,  2.48s/it, loss=0.2375, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1787/2855 [1:20:16<44:04,  2.48s/it, loss=0.1569, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1788/2855 [1:20:16<43:37,  2.45s/it, loss=0.1569, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1788/2855 [1:20:18<43:37,  2.45s/it, loss=0.2149, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1789/2855 [1:20:18<40:40,  2.29s/it, loss=0.2149, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1789/2855 [1:20:20<40:40,  2.29s/it, loss=0.2173, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1790/2855 [1:20:20<40:03,  2.26s/it, loss=0.2173, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1790/2855 [1:20:22<40:03,  2.26s/it, loss=0.2032, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1791/2855 [1:20:22<38:52,  2.19s/it, loss=0.2032, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1791/2855 [1:20:24<38:52,  2.19s/it, loss=0.1916, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1792/2855 [1:20:24<38:02,  2.15s/it, loss=0.1916, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1792/2855 [1:20:26<38:02,  2.15s/it, loss=0.1020, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1793/2855 [1:20:26<37:07,  2.10s/it, loss=0.1020, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1793/2855 [1:20:28<37:07,  2.10s/it, loss=0.2955, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1794/2855 [1:20:28<36:21,  2.06s/it, loss=0.2955, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1794/2855 [1:20:30<36:21,  2.06s/it, loss=0.4030, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1795/2855 [1:20:30<35:57,  2.03s/it, loss=0.4030, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1795/2855 [1:20:32<35:57,  2.03s/it, loss=0.2759, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1796/2855 [1:20:32<35:41,  2.02s/it, loss=0.2759, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1796/2855 [1:20:34<35:41,  2.02s/it, loss=0.1046, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1797/2855 [1:20:34<35:33,  2.02s/it, loss=0.1046, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1797/2855 [1:20:36<35:33,  2.02s/it, loss=0.0904, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1798/2855 [1:20:36<34:43,  1.97s/it, loss=0.0904, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1798/2855 [1:20:38<34:43,  1.97s/it, loss=0.2954, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1799/2855 [1:20:38<35:04,  1.99s/it, loss=0.2954, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1799/2855 [1:20:40<35:04,  1.99s/it, loss=0.1314, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1800/2855 [1:20:40<34:57,  1.99s/it, loss=0.1314, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1800/2855 [1:20:42<34:57,  1.99s/it, loss=0.3290, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1801/2855 [1:20:42<35:04,  2.00s/it, loss=0.3290, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1801/2855 [1:20:44<35:04,  2.00s/it, loss=0.1362, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1802/2855 [1:20:44<34:59,  1.99s/it, loss=0.1362, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1802/2855 [1:20:46<34:59,  1.99s/it, loss=0.3661, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1803/2855 [1:20:46<33:59,  1.94s/it, loss=0.3661, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1803/2855 [1:20:48<33:59,  1.94s/it, loss=0.2865, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1804/2855 [1:20:48<34:23,  1.96s/it, loss=0.2865, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1804/2855 [1:20:50<34:23,  1.96s/it, loss=0.2976, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1805/2855 [1:20:50<34:37,  1.98s/it, loss=0.2976, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1805/2855 [1:20:52<34:37,  1.98s/it, loss=0.1236, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1806/2855 [1:20:52<34:43,  1.99s/it, loss=0.1236, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1806/2855 [1:20:54<34:43,  1.99s/it, loss=0.2835, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1807/2855 [1:20:54<35:05,  2.01s/it, loss=0.2835, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1807/2855 [1:20:56<35:05,  2.01s/it, loss=0.2121, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1808/2855 [1:20:56<34:06,  1.95s/it, loss=0.2121, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1808/2855 [1:20:58<34:06,  1.95s/it, loss=0.2116, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1809/2855 [1:20:58<34:24,  1.97s/it, loss=0.2116, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1809/2855 [1:21:00<34:24,  1.97s/it, loss=0.2689, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1810/2855 [1:21:00<34:55,  2.00s/it, loss=0.2689, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1810/2855 [1:21:02<34:55,  2.00s/it, loss=0.0996, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1811/2855 [1:21:02<34:15,  1.97s/it, loss=0.0996, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1811/2855 [1:21:04<34:15,  1.97s/it, loss=0.1870, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1812/2855 [1:21:04<34:25,  1.98s/it, loss=0.1870, lr=2.96e-05]
Epoch 1/8:  63%|██████▎   | 1812/2855 [1:21:05<34:25,  1.98s/it, loss=0.1154, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1813/2855 [1:21:05<33:39,  1.94s/it, loss=0.1154, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1813/2855 [1:21:07<33:39,  1.94s/it, loss=0.1458, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1814/2855 [1:21:07<33:52,  1.95s/it, loss=0.1458, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1814/2855 [1:21:09<33:52,  1.95s/it, loss=0.2240, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1815/2855 [1:21:09<34:48,  2.01s/it, loss=0.2240, lr=2.96e-05]
Epoch 1/8:  64%|██████▎   | 1815/2855 [1:21:11<34:48,  2.01s/it, loss=0.1966, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1816/2855 [1:21:11<34:24,  1.99s/it, loss=0.1966, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1816/2855 [1:21:13<34:24,  1.99s/it, loss=0.1776, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1817/2855 [1:21:13<33:48,  1.95s/it, loss=0.1776, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1817/2855 [1:21:15<33:48,  1.95s/it, loss=0.2140, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1818/2855 [1:21:15<32:24,  1.88s/it, loss=0.2140, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1818/2855 [1:21:18<32:24,  1.88s/it, loss=0.2173, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1819/2855 [1:21:18<40:50,  2.36s/it, loss=0.2173, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1819/2855 [1:21:21<40:50,  2.36s/it, loss=0.4145, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1820/2855 [1:21:21<40:57,  2.37s/it, loss=0.4145, lr=2.95e-05]
Epoch 1/8:  64%|██████▎   | 1820/2855 [1:21:23<40:57,  2.37s/it, loss=0.2371, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1821/2855 [1:21:23<41:36,  2.41s/it, loss=0.2371, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1821/2855 [1:21:26<41:36,  2.41s/it, loss=0.2290, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1822/2855 [1:21:26<39:58,  2.32s/it, loss=0.2290, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1822/2855 [1:21:29<39:58,  2.32s/it, loss=0.1404, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1823/2855 [1:21:29<45:52,  2.67s/it, loss=0.1404, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1823/2855 [1:21:31<45:52,  2.67s/it, loss=0.1937, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1824/2855 [1:21:31<41:40,  2.43s/it, loss=0.1937, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1824/2855 [1:21:34<41:40,  2.43s/it, loss=0.2194, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1825/2855 [1:21:34<43:03,  2.51s/it, loss=0.2194, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1825/2855 [1:21:35<43:03,  2.51s/it, loss=0.0766, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1826/2855 [1:21:35<40:06,  2.34s/it, loss=0.0766, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1826/2855 [1:21:38<40:06,  2.34s/it, loss=0.1919, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1827/2855 [1:21:38<42:54,  2.50s/it, loss=0.1919, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1827/2855 [1:21:40<42:54,  2.50s/it, loss=0.2106, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1828/2855 [1:21:40<40:21,  2.36s/it, loss=0.2106, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1828/2855 [1:21:43<40:21,  2.36s/it, loss=0.3239, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1829/2855 [1:21:43<39:42,  2.32s/it, loss=0.3239, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1829/2855 [1:21:45<39:42,  2.32s/it, loss=0.2509, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1830/2855 [1:21:45<37:42,  2.21s/it, loss=0.2509, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1830/2855 [1:21:47<37:42,  2.21s/it, loss=0.2281, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1831/2855 [1:21:47<38:12,  2.24s/it, loss=0.2281, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1831/2855 [1:21:49<38:12,  2.24s/it, loss=0.3827, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1832/2855 [1:21:49<37:01,  2.17s/it, loss=0.3827, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1832/2855 [1:21:51<37:01,  2.17s/it, loss=0.1971, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1833/2855 [1:21:51<36:49,  2.16s/it, loss=0.1971, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1833/2855 [1:21:53<36:49,  2.16s/it, loss=0.1781, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1834/2855 [1:21:53<37:01,  2.18s/it, loss=0.1781, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1834/2855 [1:21:55<37:01,  2.18s/it, loss=0.2134, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1835/2855 [1:21:55<36:30,  2.15s/it, loss=0.2134, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1835/2855 [1:21:58<36:30,  2.15s/it, loss=0.1509, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1836/2855 [1:21:58<36:42,  2.16s/it, loss=0.1509, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1836/2855 [1:21:59<36:42,  2.16s/it, loss=0.0432, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1837/2855 [1:21:59<34:28,  2.03s/it, loss=0.0432, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1837/2855 [1:22:01<34:28,  2.03s/it, loss=0.1143, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1838/2855 [1:22:01<34:47,  2.05s/it, loss=0.1143, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1838/2855 [1:22:03<34:47,  2.05s/it, loss=0.2484, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1839/2855 [1:22:03<34:11,  2.02s/it, loss=0.2484, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1839/2855 [1:22:05<34:11,  2.02s/it, loss=0.1697, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1840/2855 [1:22:05<34:15,  2.02s/it, loss=0.1697, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1840/2855 [1:22:08<34:15,  2.02s/it, loss=0.1599, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1841/2855 [1:22:08<36:18,  2.15s/it, loss=0.1599, lr=2.95e-05]
Epoch 1/8:  64%|██████▍   | 1841/2855 [1:22:09<36:18,  2.15s/it, loss=0.2125, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1842/2855 [1:22:09<33:59,  2.01s/it, loss=0.2125, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1842/2855 [1:22:12<33:59,  2.01s/it, loss=0.3105, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1843/2855 [1:22:12<36:53,  2.19s/it, loss=0.3105, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1843/2855 [1:22:15<36:53,  2.19s/it, loss=0.1561, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1844/2855 [1:22:15<38:22,  2.28s/it, loss=0.1561, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1844/2855 [1:22:17<38:22,  2.28s/it, loss=0.1299, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1845/2855 [1:22:17<38:03,  2.26s/it, loss=0.1299, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1845/2855 [1:22:19<38:03,  2.26s/it, loss=0.2544, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1846/2855 [1:22:19<37:22,  2.22s/it, loss=0.2544, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1846/2855 [1:22:21<37:22,  2.22s/it, loss=0.1596, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1847/2855 [1:22:21<36:41,  2.18s/it, loss=0.1596, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1847/2855 [1:22:23<36:41,  2.18s/it, loss=0.2628, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1848/2855 [1:22:23<37:02,  2.21s/it, loss=0.2628, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1848/2855 [1:22:25<37:02,  2.21s/it, loss=0.2684, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1849/2855 [1:22:25<36:16,  2.16s/it, loss=0.2684, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1849/2855 [1:22:28<36:16,  2.16s/it, loss=0.2121, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1850/2855 [1:22:28<37:21,  2.23s/it, loss=0.2121, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1850/2855 [1:22:30<37:21,  2.23s/it, loss=0.1067, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1851/2855 [1:22:30<36:14,  2.17s/it, loss=0.1067, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1851/2855 [1:22:32<36:14,  2.17s/it, loss=0.3202, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1852/2855 [1:22:32<36:38,  2.19s/it, loss=0.3202, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1852/2855 [1:22:34<36:38,  2.19s/it, loss=0.2655, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1853/2855 [1:22:34<35:58,  2.15s/it, loss=0.2655, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1853/2855 [1:22:36<35:58,  2.15s/it, loss=0.1326, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1854/2855 [1:22:36<36:15,  2.17s/it, loss=0.1326, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1854/2855 [1:22:38<36:15,  2.17s/it, loss=0.2192, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1855/2855 [1:22:38<36:24,  2.18s/it, loss=0.2192, lr=2.95e-05]
Epoch 1/8:  65%|██████▍   | 1855/2855 [1:22:40<36:24,  2.18s/it, loss=0.1217, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1856/2855 [1:22:40<34:17,  2.06s/it, loss=0.1217, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1856/2855 [1:22:42<34:17,  2.06s/it, loss=0.3098, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1857/2855 [1:22:42<35:14,  2.12s/it, loss=0.3098, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1857/2855 [1:22:44<35:14,  2.12s/it, loss=0.2282, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1858/2855 [1:22:44<33:20,  2.01s/it, loss=0.2282, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1858/2855 [1:22:46<33:20,  2.01s/it, loss=0.1817, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1859/2855 [1:22:46<33:46,  2.04s/it, loss=0.1817, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1859/2855 [1:22:48<33:46,  2.04s/it, loss=0.0928, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1860/2855 [1:22:48<34:00,  2.05s/it, loss=0.0928, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1860/2855 [1:22:50<34:00,  2.05s/it, loss=0.0393, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1861/2855 [1:22:50<33:37,  2.03s/it, loss=0.0393, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1861/2855 [1:22:53<33:37,  2.03s/it, loss=0.1440, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1862/2855 [1:22:53<37:03,  2.24s/it, loss=0.1440, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1862/2855 [1:22:55<37:03,  2.24s/it, loss=0.1359, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1863/2855 [1:22:55<36:08,  2.19s/it, loss=0.1359, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1863/2855 [1:22:57<36:08,  2.19s/it, loss=0.1479, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1864/2855 [1:22:57<36:04,  2.18s/it, loss=0.1479, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1864/2855 [1:22:59<36:04,  2.18s/it, loss=0.1734, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1865/2855 [1:22:59<35:34,  2.16s/it, loss=0.1734, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1865/2855 [1:23:02<35:34,  2.16s/it, loss=0.2813, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1866/2855 [1:23:02<35:20,  2.14s/it, loss=0.2813, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1866/2855 [1:23:04<35:20,  2.14s/it, loss=0.0682, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1867/2855 [1:23:04<35:55,  2.18s/it, loss=0.0682, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1867/2855 [1:23:06<35:55,  2.18s/it, loss=0.3434, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1868/2855 [1:23:06<35:01,  2.13s/it, loss=0.3434, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1868/2855 [1:23:08<35:01,  2.13s/it, loss=0.2129, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1869/2855 [1:23:08<36:03,  2.19s/it, loss=0.2129, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1869/2855 [1:23:10<36:03,  2.19s/it, loss=0.2784, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1870/2855 [1:23:10<35:35,  2.17s/it, loss=0.2784, lr=2.95e-05]
Epoch 1/8:  65%|██████▌   | 1870/2855 [1:23:13<35:35,  2.17s/it, loss=0.0879, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1871/2855 [1:23:13<36:53,  2.25s/it, loss=0.0879, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1871/2855 [1:23:15<36:53,  2.25s/it, loss=0.2102, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1872/2855 [1:23:15<35:09,  2.15s/it, loss=0.2102, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1872/2855 [1:23:17<35:09,  2.15s/it, loss=0.2866, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1873/2855 [1:23:17<35:27,  2.17s/it, loss=0.2866, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1873/2855 [1:23:19<35:27,  2.17s/it, loss=0.1932, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1874/2855 [1:23:19<36:39,  2.24s/it, loss=0.1932, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1874/2855 [1:23:21<36:39,  2.24s/it, loss=0.3019, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1875/2855 [1:23:21<36:24,  2.23s/it, loss=0.3019, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1875/2855 [1:23:24<36:24,  2.23s/it, loss=0.2455, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1876/2855 [1:23:24<36:21,  2.23s/it, loss=0.2455, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1876/2855 [1:23:26<36:21,  2.23s/it, loss=0.2238, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1877/2855 [1:23:26<34:45,  2.13s/it, loss=0.2238, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1877/2855 [1:23:28<34:45,  2.13s/it, loss=0.0837, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1878/2855 [1:23:28<35:30,  2.18s/it, loss=0.0837, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1878/2855 [1:23:30<35:30,  2.18s/it, loss=0.1375, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1879/2855 [1:23:30<34:15,  2.11s/it, loss=0.1375, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1879/2855 [1:23:32<34:15,  2.11s/it, loss=0.1733, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1880/2855 [1:23:32<34:24,  2.12s/it, loss=0.1733, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1880/2855 [1:23:34<34:24,  2.12s/it, loss=0.0980, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1881/2855 [1:23:34<35:02,  2.16s/it, loss=0.0980, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1881/2855 [1:23:36<35:02,  2.16s/it, loss=0.1226, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1882/2855 [1:23:36<34:19,  2.12s/it, loss=0.1226, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1882/2855 [1:23:39<34:19,  2.12s/it, loss=0.1448, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1883/2855 [1:23:39<35:46,  2.21s/it, loss=0.1448, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1883/2855 [1:23:41<35:46,  2.21s/it, loss=0.1113, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1884/2855 [1:23:41<34:39,  2.14s/it, loss=0.1113, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1884/2855 [1:23:43<34:39,  2.14s/it, loss=0.2504, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1885/2855 [1:23:43<33:34,  2.08s/it, loss=0.2504, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1885/2855 [1:23:44<33:34,  2.08s/it, loss=0.2555, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1886/2855 [1:23:44<32:09,  1.99s/it, loss=0.2555, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1886/2855 [1:23:46<32:09,  1.99s/it, loss=0.1375, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1887/2855 [1:23:46<31:25,  1.95s/it, loss=0.1375, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1887/2855 [1:23:48<31:25,  1.95s/it, loss=0.2286, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1888/2855 [1:23:48<32:14,  2.00s/it, loss=0.2286, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1888/2855 [1:23:50<32:14,  2.00s/it, loss=0.0948, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1889/2855 [1:23:50<30:46,  1.91s/it, loss=0.0948, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1889/2855 [1:23:52<30:46,  1.91s/it, loss=0.1391, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1890/2855 [1:23:52<32:37,  2.03s/it, loss=0.1391, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1890/2855 [1:23:55<32:37,  2.03s/it, loss=0.0003, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1891/2855 [1:23:55<33:52,  2.11s/it, loss=0.0003, lr=2.95e-05]
Epoch 1/8:  66%|██████▌   | 1891/2855 [1:23:57<33:52,  2.11s/it, loss=0.2065, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1892/2855 [1:23:57<33:48,  2.11s/it, loss=0.2065, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1892/2855 [1:23:59<33:48,  2.11s/it, loss=0.0596, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1893/2855 [1:23:59<34:26,  2.15s/it, loss=0.0596, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1893/2855 [1:24:01<34:26,  2.15s/it, loss=0.1909, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1894/2855 [1:24:01<33:06,  2.07s/it, loss=0.1909, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1894/2855 [1:24:03<33:06,  2.07s/it, loss=0.3032, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1895/2855 [1:24:03<33:58,  2.12s/it, loss=0.3032, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1895/2855 [1:24:05<33:58,  2.12s/it, loss=0.3536, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1896/2855 [1:24:05<32:48,  2.05s/it, loss=0.3536, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1896/2855 [1:24:07<32:48,  2.05s/it, loss=0.1835, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1897/2855 [1:24:07<33:10,  2.08s/it, loss=0.1835, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1897/2855 [1:24:10<33:10,  2.08s/it, loss=0.1602, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1898/2855 [1:24:10<34:27,  2.16s/it, loss=0.1602, lr=2.95e-05]
Epoch 1/8:  66%|██████▋   | 1898/2855 [1:24:11<34:27,  2.16s/it, loss=0.2001, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1899/2855 [1:24:11<33:19,  2.09s/it, loss=0.2001, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1899/2855 [1:24:14<33:19,  2.09s/it, loss=0.0967, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1900/2855 [1:24:14<34:17,  2.15s/it, loss=0.0967, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1900/2855 [1:24:16<34:17,  2.15s/it, loss=0.3118, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1901/2855 [1:24:16<32:46,  2.06s/it, loss=0.3118, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1901/2855 [1:24:17<32:46,  2.06s/it, loss=0.2889, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1902/2855 [1:24:17<31:42,  2.00s/it, loss=0.2889, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1902/2855 [1:24:19<31:42,  2.00s/it, loss=0.0702, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1903/2855 [1:24:19<31:32,  1.99s/it, loss=0.0702, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1903/2855 [1:24:21<31:32,  1.99s/it, loss=0.1863, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1904/2855 [1:24:21<31:14,  1.97s/it, loss=0.1863, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1904/2855 [1:24:24<31:14,  1.97s/it, loss=0.2424, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1905/2855 [1:24:24<32:15,  2.04s/it, loss=0.2424, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1905/2855 [1:24:25<32:15,  2.04s/it, loss=0.1030, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1906/2855 [1:24:25<30:18,  1.92s/it, loss=0.1030, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1906/2855 [1:24:27<30:18,  1.92s/it, loss=0.1652, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1907/2855 [1:24:27<29:33,  1.87s/it, loss=0.1652, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1907/2855 [1:24:29<29:33,  1.87s/it, loss=0.0895, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1908/2855 [1:24:29<30:23,  1.93s/it, loss=0.0895, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1908/2855 [1:24:31<30:23,  1.93s/it, loss=0.0919, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1909/2855 [1:24:31<29:04,  1.84s/it, loss=0.0919, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1909/2855 [1:24:33<29:04,  1.84s/it, loss=0.1592, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1910/2855 [1:24:33<30:04,  1.91s/it, loss=0.1592, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1910/2855 [1:24:34<30:04,  1.91s/it, loss=0.2574, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1911/2855 [1:24:34<29:31,  1.88s/it, loss=0.2574, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1911/2855 [1:24:36<29:31,  1.88s/it, loss=0.3252, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1912/2855 [1:24:36<30:04,  1.91s/it, loss=0.3252, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1912/2855 [1:24:39<30:04,  1.91s/it, loss=0.1954, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1913/2855 [1:24:39<32:57,  2.10s/it, loss=0.1954, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1913/2855 [1:24:41<32:57,  2.10s/it, loss=0.1758, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1914/2855 [1:24:41<32:14,  2.06s/it, loss=0.1758, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1914/2855 [1:24:43<32:14,  2.06s/it, loss=0.0862, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1915/2855 [1:24:43<32:16,  2.06s/it, loss=0.0862, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1915/2855 [1:24:45<32:16,  2.06s/it, loss=0.1385, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1916/2855 [1:24:45<31:54,  2.04s/it, loss=0.1385, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1916/2855 [1:24:47<31:54,  2.04s/it, loss=0.4068, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1917/2855 [1:24:47<31:11,  2.00s/it, loss=0.4068, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1917/2855 [1:24:49<31:11,  2.00s/it, loss=0.2111, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1918/2855 [1:24:49<32:25,  2.08s/it, loss=0.2111, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1918/2855 [1:24:51<32:25,  2.08s/it, loss=0.3226, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1919/2855 [1:24:51<31:25,  2.01s/it, loss=0.3226, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1919/2855 [1:24:53<31:25,  2.01s/it, loss=0.0953, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1920/2855 [1:24:53<32:05,  2.06s/it, loss=0.0953, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1920/2855 [1:24:55<32:05,  2.06s/it, loss=0.2090, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1921/2855 [1:24:55<31:30,  2.02s/it, loss=0.2090, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1921/2855 [1:24:57<31:30,  2.02s/it, loss=0.1914, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1922/2855 [1:24:57<31:13,  2.01s/it, loss=0.1914, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1922/2855 [1:24:59<31:13,  2.01s/it, loss=0.1168, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1923/2855 [1:24:59<32:32,  2.10s/it, loss=0.1168, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1923/2855 [1:25:01<32:32,  2.10s/it, loss=0.1336, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1924/2855 [1:25:01<31:17,  2.02s/it, loss=0.1336, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1924/2855 [1:25:03<31:17,  2.02s/it, loss=0.2522, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1925/2855 [1:25:03<31:36,  2.04s/it, loss=0.2522, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1925/2855 [1:25:05<31:36,  2.04s/it, loss=0.0016, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1926/2855 [1:25:05<29:50,  1.93s/it, loss=0.0016, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1926/2855 [1:25:07<29:50,  1.93s/it, loss=0.1895, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1927/2855 [1:25:07<31:32,  2.04s/it, loss=0.1895, lr=2.95e-05]
Epoch 1/8:  67%|██████▋   | 1927/2855 [1:25:10<31:32,  2.04s/it, loss=0.2730, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1928/2855 [1:25:10<32:42,  2.12s/it, loss=0.2730, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1928/2855 [1:25:11<32:42,  2.12s/it, loss=0.1942, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1929/2855 [1:25:11<31:21,  2.03s/it, loss=0.1942, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1929/2855 [1:25:14<31:21,  2.03s/it, loss=0.1295, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1930/2855 [1:25:14<31:58,  2.07s/it, loss=0.1295, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1930/2855 [1:25:15<31:58,  2.07s/it, loss=0.3658, lr=2.95e-05]
Epoch 1/8:  68%|█��████▊   | 1931/2855 [1:25:15<30:40,  1.99s/it, loss=0.3658, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1931/2855 [1:25:17<30:40,  1.99s/it, loss=0.1043, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1932/2855 [1:25:17<30:19,  1.97s/it, loss=0.1043, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1932/2855 [1:25:20<30:19,  1.97s/it, loss=0.1352, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1933/2855 [1:25:20<32:00,  2.08s/it, loss=0.1352, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1933/2855 [1:25:22<32:00,  2.08s/it, loss=0.2614, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1934/2855 [1:25:22<30:55,  2.01s/it, loss=0.2614, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1934/2855 [1:25:24<30:55,  2.01s/it, loss=0.2492, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1935/2855 [1:25:24<32:06,  2.09s/it, loss=0.2492, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1935/2855 [1:25:26<32:06,  2.09s/it, loss=0.3543, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1936/2855 [1:25:26<31:40,  2.07s/it, loss=0.3543, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1936/2855 [1:25:28<31:40,  2.07s/it, loss=0.1699, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1937/2855 [1:25:28<31:22,  2.05s/it, loss=0.1699, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1937/2855 [1:25:30<31:22,  2.05s/it, loss=0.1348, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1938/2855 [1:25:30<31:26,  2.06s/it, loss=0.1348, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1938/2855 [1:25:32<31:26,  2.06s/it, loss=0.2021, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1939/2855 [1:25:32<29:18,  1.92s/it, loss=0.2021, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1939/2855 [1:25:34<29:18,  1.92s/it, loss=0.2452, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1940/2855 [1:25:34<29:54,  1.96s/it, loss=0.2452, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1940/2855 [1:25:35<29:54,  1.96s/it, loss=0.2102, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1941/2855 [1:25:35<28:52,  1.90s/it, loss=0.2102, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1941/2855 [1:25:37<28:52,  1.90s/it, loss=0.2158, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1942/2855 [1:25:37<29:21,  1.93s/it, loss=0.2158, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1942/2855 [1:25:40<29:21,  1.93s/it, loss=0.2218, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1943/2855 [1:25:40<30:29,  2.01s/it, loss=0.2218, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1943/2855 [1:25:41<30:29,  2.01s/it, loss=0.1139, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1944/2855 [1:25:41<29:24,  1.94s/it, loss=0.1139, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1944/2855 [1:25:44<29:24,  1.94s/it, loss=0.3331, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1945/2855 [1:25:44<31:54,  2.10s/it, loss=0.3331, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1945/2855 [1:25:46<31:54,  2.10s/it, loss=0.0002, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1946/2855 [1:25:46<30:44,  2.03s/it, loss=0.0002, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1946/2855 [1:25:48<30:44,  2.03s/it, loss=0.1895, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1947/2855 [1:25:48<30:26,  2.01s/it, loss=0.1895, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1947/2855 [1:25:50<30:26,  2.01s/it, loss=0.2435, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1948/2855 [1:25:50<31:19,  2.07s/it, loss=0.2435, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1948/2855 [1:25:52<31:19,  2.07s/it, loss=0.2503, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1949/2855 [1:25:52<29:48,  1.97s/it, loss=0.2503, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1949/2855 [1:25:54<29:48,  1.97s/it, loss=0.2623, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1950/2855 [1:25:54<31:09,  2.07s/it, loss=0.2623, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1950/2855 [1:25:56<31:09,  2.07s/it, loss=0.1355, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1951/2855 [1:25:56<30:26,  2.02s/it, loss=0.1355, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1951/2855 [1:25:58<30:26,  2.02s/it, loss=0.2544, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1952/2855 [1:25:58<29:36,  1.97s/it, loss=0.2544, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1952/2855 [1:26:00<29:36,  1.97s/it, loss=0.2921, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1953/2855 [1:26:00<30:52,  2.05s/it, loss=0.2921, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1953/2855 [1:26:02<30:52,  2.05s/it, loss=0.2054, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1954/2855 [1:26:02<29:44,  1.98s/it, loss=0.2054, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1954/2855 [1:26:04<29:44,  1.98s/it, loss=0.0687, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1955/2855 [1:26:04<30:54,  2.06s/it, loss=0.0687, lr=2.95e-05]
Epoch 1/8:  68%|██████▊   | 1955/2855 [1:26:06<30:54,  2.06s/it, loss=0.3190, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1956/2855 [1:26:06<29:41,  1.98s/it, loss=0.3190, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1956/2855 [1:26:07<29:41,  1.98s/it, loss=0.1064, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1957/2855 [1:26:07<28:09,  1.88s/it, loss=0.1064, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1957/2855 [1:26:09<28:09,  1.88s/it, loss=0.0753, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1958/2855 [1:26:09<28:53,  1.93s/it, loss=0.0753, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1958/2855 [1:26:11<28:53,  1.93s/it, loss=0.1550, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1959/2855 [1:26:11<28:09,  1.89s/it, loss=0.1550, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1959/2855 [1:26:13<28:09,  1.89s/it, loss=0.1344, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1960/2855 [1:26:13<28:58,  1.94s/it, loss=0.1344, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1960/2855 [1:26:15<28:58,  1.94s/it, loss=0.2037, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1961/2855 [1:26:15<28:54,  1.94s/it, loss=0.2037, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1961/2855 [1:26:17<28:54,  1.94s/it, loss=0.4028, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1962/2855 [1:26:17<27:13,  1.83s/it, loss=0.4028, lr=2.95e-05]
Epoch 1/8:  69%|██████▊   | 1962/2855 [1:26:19<27:13,  1.83s/it, loss=0.0576, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1963/2855 [1:26:19<29:55,  2.01s/it, loss=0.0576, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1963/2855 [1:26:21<29:55,  2.01s/it, loss=0.1871, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1964/2855 [1:26:21<29:39,  2.00s/it, loss=0.1871, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1964/2855 [1:26:23<29:39,  2.00s/it, loss=0.2511, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1965/2855 [1:26:23<29:16,  1.97s/it, loss=0.2511, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1965/2855 [1:26:25<29:16,  1.97s/it, loss=0.3440, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1966/2855 [1:26:25<29:50,  2.01s/it, loss=0.3440, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1966/2855 [1:26:27<29:50,  2.01s/it, loss=0.1436, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1967/2855 [1:26:27<29:09,  1.97s/it, loss=0.1436, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1967/2855 [1:26:29<29:09,  1.97s/it, loss=0.1195, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1968/2855 [1:26:29<30:20,  2.05s/it, loss=0.1195, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1968/2855 [1:26:31<30:20,  2.05s/it, loss=0.2474, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1969/2855 [1:26:31<29:18,  1.98s/it, loss=0.2474, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1969/2855 [1:26:33<29:18,  1.98s/it, loss=0.0984, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1970/2855 [1:26:33<29:11,  1.98s/it, loss=0.0984, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1970/2855 [1:26:35<29:11,  1.98s/it, loss=0.1166, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1971/2855 [1:26:35<29:57,  2.03s/it, loss=0.1166, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1971/2855 [1:26:37<29:57,  2.03s/it, loss=0.1165, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1972/2855 [1:26:37<29:01,  1.97s/it, loss=0.1165, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1972/2855 [1:26:39<29:01,  1.97s/it, loss=0.1736, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1973/2855 [1:26:39<30:22,  2.07s/it, loss=0.1736, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1973/2855 [1:26:41<30:22,  2.07s/it, loss=0.3003, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1974/2855 [1:26:41<29:24,  2.00s/it, loss=0.3003, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1974/2855 [1:26:43<29:24,  2.00s/it, loss=0.2761, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1975/2855 [1:26:43<28:11,  1.92s/it, loss=0.2761, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1975/2855 [1:26:45<28:11,  1.92s/it, loss=0.2548, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1976/2855 [1:26:45<28:42,  1.96s/it, loss=0.2548, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1976/2855 [1:26:47<28:42,  1.96s/it, loss=0.2844, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1977/2855 [1:26:47<29:02,  1.98s/it, loss=0.2844, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1977/2855 [1:26:49<29:02,  1.98s/it, loss=0.0369, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1978/2855 [1:26:49<30:54,  2.11s/it, loss=0.0369, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1978/2855 [1:26:51<30:54,  2.11s/it, loss=0.3063, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1979/2855 [1:26:51<29:45,  2.04s/it, loss=0.3063, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1979/2855 [1:26:53<29:45,  2.04s/it, loss=0.2266, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1980/2855 [1:26:53<28:52,  1.98s/it, loss=0.2266, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1980/2855 [1:26:55<28:52,  1.98s/it, loss=0.1546, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1981/2855 [1:26:55<30:12,  2.07s/it, loss=0.1546, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1981/2855 [1:26:57<30:12,  2.07s/it, loss=0.1749, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1982/2855 [1:26:57<28:47,  1.98s/it, loss=0.1749, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1982/2855 [1:26:59<28:47,  1.98s/it, loss=0.1281, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1983/2855 [1:26:59<29:55,  2.06s/it, loss=0.1281, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1983/2855 [1:27:01<29:55,  2.06s/it, loss=0.3234, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1984/2855 [1:27:01<28:55,  1.99s/it, loss=0.3234, lr=2.95e-05]
Epoch 1/8:  69%|██████▉   | 1984/2855 [1:27:03<28:55,  1.99s/it, loss=0.1035, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1985/2855 [1:27:03<28:14,  1.95s/it, loss=0.1035, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1985/2855 [1:27:05<28:14,  1.95s/it, loss=0.1675, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1986/2855 [1:27:05<29:40,  2.05s/it, loss=0.1675, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1986/2855 [1:27:07<29:40,  2.05s/it, loss=0.2937, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1987/2855 [1:27:07<28:38,  1.98s/it, loss=0.2937, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1987/2855 [1:27:09<28:38,  1.98s/it, loss=0.1557, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1988/2855 [1:27:09<29:22,  2.03s/it, loss=0.1557, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1988/2855 [1:27:11<29:22,  2.03s/it, loss=0.2566, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1989/2855 [1:27:11<28:07,  1.95s/it, loss=0.2566, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1989/2855 [1:27:13<28:07,  1.95s/it, loss=0.2884, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1990/2855 [1:27:13<27:09,  1.88s/it, loss=0.2884, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1990/2855 [1:27:15<27:09,  1.88s/it, loss=0.1311, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1991/2855 [1:27:15<29:01,  2.02s/it, loss=0.1311, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1991/2855 [1:27:17<29:01,  2.02s/it, loss=0.1130, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1992/2855 [1:27:17<27:06,  1.88s/it, loss=0.1130, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1992/2855 [1:27:18<27:06,  1.88s/it, loss=0.3144, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1993/2855 [1:27:18<26:05,  1.82s/it, loss=0.3144, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1993/2855 [1:27:21<26:05,  1.82s/it, loss=0.0894, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1994/2855 [1:27:21<27:30,  1.92s/it, loss=0.0894, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1994/2855 [1:27:22<27:30,  1.92s/it, loss=0.2792, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1995/2855 [1:27:22<25:56,  1.81s/it, loss=0.2792, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1995/2855 [1:27:24<25:56,  1.81s/it, loss=0.1457, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1996/2855 [1:27:24<27:54,  1.95s/it, loss=0.1457, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1996/2855 [1:27:26<27:54,  1.95s/it, loss=0.3874, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1997/2855 [1:27:26<28:16,  1.98s/it, loss=0.3874, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1997/2855 [1:27:28<28:16,  1.98s/it, loss=0.1598, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1998/2855 [1:27:28<28:18,  1.98s/it, loss=0.1598, lr=2.95e-05]
Epoch 1/8:  70%|██████▉   | 1998/2855 [1:27:31<28:18,  1.98s/it, loss=0.1497, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 1999/2855 [1:27:31<29:07,  2.04s/it, loss=0.1497, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 1999/2855 [1:27:33<29:07,  2.04s/it, loss=0.2977, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2000/2855 [1:27:33<28:17,  1.99s/it, loss=0.2977, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2000/2855 [1:27:35<28:17,  1.99s/it, loss=0.2078, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2001/2855 [1:27:35<30:06,  2.11s/it, loss=0.2078, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2001/2855 [1:27:37<30:06,  2.11s/it, loss=0.1338, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2002/2855 [1:27:37<29:01,  2.04s/it, loss=0.1338, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2002/2855 [1:27:39<29:01,  2.04s/it, loss=0.1253, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2003/2855 [1:27:39<28:32,  2.01s/it, loss=0.1253, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2003/2855 [1:27:41<28:32,  2.01s/it, loss=0.3463, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2004/2855 [1:27:41<29:40,  2.09s/it, loss=0.3463, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2004/2855 [1:27:43<29:40,  2.09s/it, loss=0.1735, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2005/2855 [1:27:43<28:40,  2.02s/it, loss=0.1735, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2005/2855 [1:27:45<28:40,  2.02s/it, loss=0.2296, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2006/2855 [1:27:45<29:21,  2.07s/it, loss=0.2296, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2006/2855 [1:27:47<29:21,  2.07s/it, loss=0.2152, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2007/2855 [1:27:47<28:20,  2.01s/it, loss=0.2152, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2007/2855 [1:27:49<28:20,  2.01s/it, loss=0.1856, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2008/2855 [1:27:49<28:03,  1.99s/it, loss=0.1856, lr=2.95e-05]
Epoch 1/8:  70%|███████   | 2008/2855 [1:27:51<28:03,  1.99s/it, loss=0.1905, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2009/2855 [1:27:51<27:42,  1.96s/it, loss=0.1905, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2009/2855 [1:27:52<27:42,  1.96s/it, loss=0.0750, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2010/2855 [1:27:52<25:42,  1.83s/it, loss=0.0750, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2010/2855 [1:27:54<25:42,  1.83s/it, loss=0.1318, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2011/2855 [1:27:54<26:37,  1.89s/it, loss=0.1318, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2011/2855 [1:27:56<26:37,  1.89s/it, loss=0.0678, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2012/2855 [1:27:56<26:38,  1.90s/it, loss=0.0678, lr=2.94e-05]
Epoch 1/8:  70%|███████   | 2012/2855 [1:27:58<26:38,  1.90s/it, loss=0.1621, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2013/2855 [1:27:58<25:19,  1.80s/it, loss=0.1621, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2013/2855 [1:28:00<25:19,  1.80s/it, loss=0.4889, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2014/2855 [1:28:00<27:38,  1.97s/it, loss=0.4889, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2014/2855 [1:28:02<27:38,  1.97s/it, loss=0.1867, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2015/2855 [1:28:02<28:03,  2.00s/it, loss=0.1867, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2015/2855 [1:28:04<28:03,  2.00s/it, loss=0.1006, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2016/2855 [1:28:04<28:17,  2.02s/it, loss=0.1006, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2016/2855 [1:28:06<28:17,  2.02s/it, loss=0.2954, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2017/2855 [1:28:06<28:21,  2.03s/it, loss=0.2954, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2017/2855 [1:28:08<28:21,  2.03s/it, loss=0.4783, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2018/2855 [1:28:08<27:11,  1.95s/it, loss=0.4783, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2018/2855 [1:28:11<27:11,  1.95s/it, loss=0.2639, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2019/2855 [1:28:11<28:51,  2.07s/it, loss=0.2639, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2019/2855 [1:28:12<28:51,  2.07s/it, loss=0.2710, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2020/2855 [1:28:12<28:09,  2.02s/it, loss=0.2710, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2020/2855 [1:28:15<28:09,  2.02s/it, loss=0.1924, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2021/2855 [1:28:15<29:46,  2.14s/it, loss=0.1924, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2021/2855 [1:28:17<29:46,  2.14s/it, loss=0.1231, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2022/2855 [1:28:17<28:54,  2.08s/it, loss=0.1231, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2022/2855 [1:28:19<28:54,  2.08s/it, loss=0.2944, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2023/2855 [1:28:19<27:54,  2.01s/it, loss=0.2944, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2023/2855 [1:28:21<27:54,  2.01s/it, loss=0.1737, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2024/2855 [1:28:21<29:28,  2.13s/it, loss=0.1737, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2024/2855 [1:28:23<29:28,  2.13s/it, loss=0.1824, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2025/2855 [1:28:23<28:22,  2.05s/it, loss=0.1824, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2025/2855 [1:28:25<28:22,  2.05s/it, loss=0.0681, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2026/2855 [1:28:25<28:14,  2.04s/it, loss=0.0681, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2026/2855 [1:28:27<28:14,  2.04s/it, loss=0.1162, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2027/2855 [1:28:27<26:30,  1.92s/it, loss=0.1162, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2027/2855 [1:28:28<26:30,  1.92s/it, loss=0.1716, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2028/2855 [1:28:28<25:03,  1.82s/it, loss=0.1716, lr=2.94e-05]
Epoch 1/8:  71%|███���███   | 2028/2855 [1:28:31<25:03,  1.82s/it, loss=0.2454, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2029/2855 [1:28:31<28:44,  2.09s/it, loss=0.2454, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2029/2855 [1:28:33<28:44,  2.09s/it, loss=0.0028, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2030/2855 [1:28:33<27:54,  2.03s/it, loss=0.0028, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2030/2855 [1:28:35<27:54,  2.03s/it, loss=0.0915, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2031/2855 [1:28:35<28:03,  2.04s/it, loss=0.0915, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2031/2855 [1:28:37<28:03,  2.04s/it, loss=0.2078, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2032/2855 [1:28:37<28:43,  2.09s/it, loss=0.2078, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2032/2855 [1:28:39<28:43,  2.09s/it, loss=0.0394, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2033/2855 [1:28:39<28:07,  2.05s/it, loss=0.0394, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2033/2855 [1:28:41<28:07,  2.05s/it, loss=0.1319, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2034/2855 [1:28:41<28:56,  2.12s/it, loss=0.1319, lr=2.94e-05]
Epoch 1/8:  71%|███████   | 2034/2855 [1:28:43<28:56,  2.12s/it, loss=0.2740, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2035/2855 [1:28:43<27:41,  2.03s/it, loss=0.2740, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2035/2855 [1:28:45<27:41,  2.03s/it, loss=0.2075, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2036/2855 [1:28:45<28:39,  2.10s/it, loss=0.2075, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2036/2855 [1:28:47<28:39,  2.10s/it, loss=0.2088, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2037/2855 [1:28:47<27:38,  2.03s/it, loss=0.2088, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2037/2855 [1:28:49<27:38,  2.03s/it, loss=0.2679, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2038/2855 [1:28:49<26:49,  1.97s/it, loss=0.2679, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2038/2855 [1:28:52<26:49,  1.97s/it, loss=0.0979, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2039/2855 [1:28:52<28:47,  2.12s/it, loss=0.0979, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2039/2855 [1:28:53<28:47,  2.12s/it, loss=0.3219, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2040/2855 [1:28:53<27:26,  2.02s/it, loss=0.3219, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2040/2855 [1:28:55<27:26,  2.02s/it, loss=0.4991, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2041/2855 [1:28:55<27:42,  2.04s/it, loss=0.4991, lr=2.94e-05]
Epoch 1/8:  71%|███████▏  | 2041/2855 [1:28:57<27:42,  2.04s/it, loss=0.1029, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2042/2855 [1:28:57<26:15,  1.94s/it, loss=0.1029, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2042/2855 [1:28:59<26:15,  1.94s/it, loss=0.0821, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2043/2855 [1:28:59<24:41,  1.82s/it, loss=0.0821, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2043/2855 [1:29:01<24:41,  1.82s/it, loss=0.2495, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2044/2855 [1:29:01<26:29,  1.96s/it, loss=0.2495, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2044/2855 [1:29:03<26:29,  1.96s/it, loss=0.1469, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2045/2855 [1:29:03<27:07,  2.01s/it, loss=0.1469, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2045/2855 [1:29:05<27:07,  2.01s/it, loss=0.1088, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2046/2855 [1:29:05<28:51,  2.14s/it, loss=0.1088, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2046/2855 [1:29:07<28:51,  2.14s/it, loss=0.2465, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2047/2855 [1:29:07<27:44,  2.06s/it, loss=0.2465, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2047/2855 [1:29:09<27:44,  2.06s/it, loss=0.0962, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2048/2855 [1:29:09<27:13,  2.02s/it, loss=0.0962, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2048/2855 [1:29:12<27:13,  2.02s/it, loss=0.2478, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2049/2855 [1:29:12<28:13,  2.10s/it, loss=0.2478, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2049/2855 [1:29:13<28:13,  2.10s/it, loss=0.3425, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2050/2855 [1:29:13<27:11,  2.03s/it, loss=0.3425, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2050/2855 [1:29:16<27:11,  2.03s/it, loss=0.1261, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2051/2855 [1:29:16<28:00,  2.09s/it, loss=0.1261, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2051/2855 [1:29:18<28:00,  2.09s/it, loss=0.3488, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2052/2855 [1:29:18<27:10,  2.03s/it, loss=0.3488, lr=2.94e-05]
Epoch 1/8:  72%|██████��▏  | 2052/2855 [1:29:19<27:10,  2.03s/it, loss=0.1967, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2053/2855 [1:29:19<26:37,  1.99s/it, loss=0.1967, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2053/2855 [1:29:22<26:37,  1.99s/it, loss=0.2003, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2054/2855 [1:29:22<27:40,  2.07s/it, loss=0.2003, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2054/2855 [1:29:24<27:40,  2.07s/it, loss=0.0784, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2055/2855 [1:29:24<26:27,  1.98s/it, loss=0.0784, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2055/2855 [1:29:26<26:27,  1.98s/it, loss=0.0459, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2056/2855 [1:29:26<27:13,  2.04s/it, loss=0.0459, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2056/2855 [1:29:27<27:13,  2.04s/it, loss=0.1952, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2057/2855 [1:29:27<25:46,  1.94s/it, loss=0.1952, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2057/2855 [1:29:29<25:46,  1.94s/it, loss=0.2320, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2058/2855 [1:29:29<24:00,  1.81s/it, loss=0.2320, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2058/2855 [1:29:31<24:00,  1.81s/it, loss=0.1772, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2059/2855 [1:29:31<25:32,  1.92s/it, loss=0.1772, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2059/2855 [1:29:33<25:32,  1.92s/it, loss=0.3012, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2060/2855 [1:29:33<24:41,  1.86s/it, loss=0.3012, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2060/2855 [1:29:35<24:41,  1.86s/it, loss=0.3340, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2061/2855 [1:29:35<24:58,  1.89s/it, loss=0.3340, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2061/2855 [1:29:37<24:58,  1.89s/it, loss=0.0957, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2062/2855 [1:29:37<25:21,  1.92s/it, loss=0.0957, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2062/2855 [1:29:39<25:21,  1.92s/it, loss=0.1653, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2063/2855 [1:29:39<25:03,  1.90s/it, loss=0.1653, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2063/2855 [1:29:41<25:03,  1.90s/it, loss=0.1363, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2064/2855 [1:29:41<28:33,  2.17s/it, loss=0.1363, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2064/2855 [1:29:43<28:33,  2.17s/it, loss=0.3224, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2065/2855 [1:29:43<27:57,  2.12s/it, loss=0.3224, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2065/2855 [1:29:45<27:57,  2.12s/it, loss=0.1700, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2066/2855 [1:29:45<27:45,  2.11s/it, loss=0.1700, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2066/2855 [1:29:47<27:45,  2.11s/it, loss=0.2318, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2067/2855 [1:29:47<27:14,  2.07s/it, loss=0.2318, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2067/2855 [1:29:49<27:14,  2.07s/it, loss=0.0701, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2068/2855 [1:29:49<25:54,  1.98s/it, loss=0.0701, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2068/2855 [1:29:52<25:54,  1.98s/it, loss=0.1583, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2069/2855 [1:29:52<27:16,  2.08s/it, loss=0.1583, lr=2.94e-05]
Epoch 1/8:  72%|███████▏  | 2069/2855 [1:29:53<27:16,  2.08s/it, loss=0.1087, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2070/2855 [1:29:53<26:09,  2.00s/it, loss=0.1087, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2070/2855 [1:29:55<26:09,  2.00s/it, loss=0.2293, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2071/2855 [1:29:55<26:35,  2.03s/it, loss=0.2293, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2071/2855 [1:29:57<26:35,  2.03s/it, loss=0.1327, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2072/2855 [1:29:57<26:19,  2.02s/it, loss=0.1327, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2072/2855 [1:29:59<26:19,  2.02s/it, loss=0.2417, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2073/2855 [1:29:59<25:09,  1.93s/it, loss=0.2417, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2073/2855 [1:30:02<25:09,  1.93s/it, loss=0.1944, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2074/2855 [1:30:02<26:46,  2.06s/it, loss=0.1944, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2074/2855 [1:30:03<26:46,  2.06s/it, loss=0.2229, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2075/2855 [1:30:03<25:28,  1.96s/it, loss=0.2229, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2075/2855 [1:30:05<25:28,  1.96s/it, loss=0.2322, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2076/2855 [1:30:05<24:59,  1.92s/it, loss=0.2322, lr=2.94e-05]
Epoch 1/8:  73%|█���█████▎  | 2076/2855 [1:30:07<24:59,  1.92s/it, loss=0.2703, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2077/2855 [1:30:07<26:01,  2.01s/it, loss=0.2703, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2077/2855 [1:30:09<26:01,  2.01s/it, loss=0.0773, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2078/2855 [1:30:09<25:34,  1.97s/it, loss=0.0773, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2078/2855 [1:30:12<25:34,  1.97s/it, loss=0.2512, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2079/2855 [1:30:12<27:05,  2.09s/it, loss=0.2512, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2079/2855 [1:30:13<27:05,  2.09s/it, loss=0.2217, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2080/2855 [1:30:13<26:05,  2.02s/it, loss=0.2217, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2080/2855 [1:30:15<26:05,  2.02s/it, loss=0.2297, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2081/2855 [1:30:15<26:06,  2.02s/it, loss=0.2297, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2081/2855 [1:30:18<26:06,  2.02s/it, loss=0.1151, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2082/2855 [1:30:18<26:10,  2.03s/it, loss=0.1151, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2082/2855 [1:30:19<26:10,  2.03s/it, loss=0.0747, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2083/2855 [1:30:19<25:02,  1.95s/it, loss=0.0747, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2083/2855 [1:30:22<25:02,  1.95s/it, loss=0.3084, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2084/2855 [1:30:22<26:12,  2.04s/it, loss=0.3084, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2084/2855 [1:30:23<26:12,  2.04s/it, loss=0.1478, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2085/2855 [1:30:23<25:29,  1.99s/it, loss=0.1478, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2085/2855 [1:30:25<25:29,  1.99s/it, loss=0.2427, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2086/2855 [1:30:25<25:25,  1.98s/it, loss=0.2427, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2086/2855 [1:30:27<25:25,  1.98s/it, loss=0.2720, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2087/2855 [1:30:27<26:01,  2.03s/it, loss=0.2720, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2087/2855 [1:30:29<26:01,  2.03s/it, loss=0.1820, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2088/2855 [1:30:29<24:53,  1.95s/it, loss=0.1820, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2088/2855 [1:30:31<24:53,  1.95s/it, loss=0.1417, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2089/2855 [1:30:31<25:24,  1.99s/it, loss=0.1417, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2089/2855 [1:30:33<25:24,  1.99s/it, loss=0.1368, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2090/2855 [1:30:33<24:09,  1.89s/it, loss=0.1368, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2090/2855 [1:30:34<24:09,  1.89s/it, loss=0.2201, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2091/2855 [1:30:35<22:35,  1.77s/it, loss=0.2201, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2091/2855 [1:30:37<22:35,  1.77s/it, loss=0.1586, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2092/2855 [1:30:37<24:34,  1.93s/it, loss=0.1586, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2092/2855 [1:30:38<24:34,  1.93s/it, loss=0.1110, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2093/2855 [1:30:38<23:37,  1.86s/it, loss=0.1110, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2093/2855 [1:30:40<23:37,  1.86s/it, loss=0.1370, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2094/2855 [1:30:40<23:20,  1.84s/it, loss=0.1370, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2094/2855 [1:30:42<23:20,  1.84s/it, loss=0.2771, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2095/2855 [1:30:42<24:03,  1.90s/it, loss=0.2771, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2095/2855 [1:30:44<24:03,  1.90s/it, loss=0.1964, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2096/2855 [1:30:44<22:41,  1.79s/it, loss=0.1964, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2096/2855 [1:30:46<22:41,  1.79s/it, loss=0.2589, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2097/2855 [1:30:46<25:25,  2.01s/it, loss=0.2589, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2097/2855 [1:30:49<25:25,  2.01s/it, loss=0.2094, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2098/2855 [1:30:49<25:55,  2.06s/it, loss=0.2094, lr=2.94e-05]
Epoch 1/8:  73%|███████▎  | 2098/2855 [1:30:51<25:55,  2.06s/it, loss=0.1021, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2099/2855 [1:30:51<25:58,  2.06s/it, loss=0.1021, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2099/2855 [1:30:53<25:58,  2.06s/it, loss=0.2372, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2100/2855 [1:30:53<26:14,  2.09s/it, loss=0.2372, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2100/2855 [1:30:55<26:14,  2.09s/it, loss=0.2494, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2101/2855 [1:30:55<25:10,  2.00s/it, loss=0.2494, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2101/2855 [1:30:57<25:10,  2.00s/it, loss=0.0985, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2102/2855 [1:30:57<26:14,  2.09s/it, loss=0.0985, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2102/2855 [1:30:59<26:14,  2.09s/it, loss=0.2062, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2103/2855 [1:30:59<25:16,  2.02s/it, loss=0.2062, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2103/2855 [1:31:01<25:16,  2.02s/it, loss=0.1268, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2104/2855 [1:31:01<25:33,  2.04s/it, loss=0.1268, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2104/2855 [1:31:03<25:33,  2.04s/it, loss=0.1377, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2105/2855 [1:31:03<25:40,  2.05s/it, loss=0.1377, lr=2.94e-05]
Epoch 1/8:  74%|███████▎  | 2105/2855 [1:31:05<25:40,  2.05s/it, loss=0.1715, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2106/2855 [1:31:05<24:35,  1.97s/it, loss=0.1715, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2106/2855 [1:31:07<24:35,  1.97s/it, loss=0.2601, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2107/2855 [1:31:07<25:48,  2.07s/it, loss=0.2601, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2107/2855 [1:31:09<25:48,  2.07s/it, loss=0.0816, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2108/2855 [1:31:09<25:08,  2.02s/it, loss=0.0816, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2108/2855 [1:31:11<25:08,  2.02s/it, loss=0.1923, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2109/2855 [1:31:11<25:23,  2.04s/it, loss=0.1923, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2109/2855 [1:31:13<25:23,  2.04s/it, loss=0.2460, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2110/2855 [1:31:13<25:27,  2.05s/it, loss=0.2460, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2110/2855 [1:31:15<25:27,  2.05s/it, loss=0.1981, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2111/2855 [1:31:15<24:44,  2.00s/it, loss=0.1981, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2111/2855 [1:31:17<24:44,  2.00s/it, loss=0.1510, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2112/2855 [1:31:17<25:33,  2.06s/it, loss=0.1510, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2112/2855 [1:31:19<25:33,  2.06s/it, loss=0.1456, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2113/2855 [1:31:19<24:44,  2.00s/it, loss=0.1456, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2113/2855 [1:31:21<24:44,  2.00s/it, loss=0.2336, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2114/2855 [1:31:21<25:20,  2.05s/it, loss=0.2336, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2114/2855 [1:31:23<25:20,  2.05s/it, loss=0.1420, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2115/2855 [1:31:23<25:11,  2.04s/it, loss=0.1420, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2115/2855 [1:31:25<25:11,  2.04s/it, loss=0.1694, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2116/2855 [1:31:25<24:05,  1.96s/it, loss=0.1694, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2116/2855 [1:31:27<24:05,  1.96s/it, loss=0.3453, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2117/2855 [1:31:27<25:22,  2.06s/it, loss=0.3453, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2117/2855 [1:31:29<25:22,  2.06s/it, loss=0.2208, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2118/2855 [1:31:29<24:46,  2.02s/it, loss=0.2208, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2118/2855 [1:31:31<24:46,  2.02s/it, loss=0.1943, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2119/2855 [1:31:31<24:56,  2.03s/it, loss=0.1943, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2119/2855 [1:31:33<24:56,  2.03s/it, loss=0.3860, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2120/2855 [1:31:33<25:13,  2.06s/it, loss=0.3860, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2120/2855 [1:31:35<25:13,  2.06s/it, loss=0.0418, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2121/2855 [1:31:35<23:56,  1.96s/it, loss=0.0418, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2121/2855 [1:31:37<23:56,  1.96s/it, loss=0.0610, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2122/2855 [1:31:37<24:16,  1.99s/it, loss=0.0610, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2122/2855 [1:31:39<24:16,  1.99s/it, loss=0.3760, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2123/2855 [1:31:39<23:36,  1.94s/it, loss=0.3760, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2123/2855 [1:31:42<23:36,  1.94s/it, loss=0.1322, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2124/2855 [1:31:42<28:31,  2.34s/it, loss=0.1322, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2124/2855 [1:31:44<28:31,  2.34s/it, loss=0.1894, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2125/2855 [1:31:44<27:51,  2.29s/it, loss=0.1894, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2125/2855 [1:31:47<27:51,  2.29s/it, loss=0.1013, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2126/2855 [1:31:47<27:18,  2.25s/it, loss=0.1013, lr=2.94e-05]
Epoch 1/8:  74%|███████▍  | 2126/2855 [1:31:49<27:18,  2.25s/it, loss=0.2128, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2127/2855 [1:31:49<26:25,  2.18s/it, loss=0.2128, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2127/2855 [1:31:51<26:25,  2.18s/it, loss=0.2686, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2128/2855 [1:31:51<26:45,  2.21s/it, loss=0.2686, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2128/2855 [1:31:54<26:45,  2.21s/it, loss=0.2398, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2129/2855 [1:31:54<28:43,  2.37s/it, loss=0.2398, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2129/2855 [1:31:56<28:43,  2.37s/it, loss=0.1691, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2130/2855 [1:31:56<29:05,  2.41s/it, loss=0.1691, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2130/2855 [1:32:00<29:05,  2.41s/it, loss=0.3997, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2131/2855 [1:32:00<36:07,  2.99s/it, loss=0.3997, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2131/2855 [1:32:04<36:07,  2.99s/it, loss=0.2374, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2132/2855 [1:32:04<38:26,  3.19s/it, loss=0.2374, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2132/2855 [1:32:10<38:26,  3.19s/it, loss=0.0376, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2133/2855 [1:32:10<49:24,  4.11s/it, loss=0.0376, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2133/2855 [1:32:13<49:24,  4.11s/it, loss=0.2051, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2134/2855 [1:32:13<43:17,  3.60s/it, loss=0.2051, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2134/2855 [1:32:15<43:17,  3.60s/it, loss=0.1919, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2135/2855 [1:32:15<37:36,  3.13s/it, loss=0.1919, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2135/2855 [1:32:17<37:36,  3.13s/it, loss=0.1849, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2136/2855 [1:32:17<33:02,  2.76s/it, loss=0.1849, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2136/2855 [1:32:19<33:02,  2.76s/it, loss=0.2708, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2137/2855 [1:32:19<29:42,  2.48s/it, loss=0.2708, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2137/2855 [1:32:21<29:42,  2.48s/it, loss=0.3327, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2138/2855 [1:32:21<28:28,  2.38s/it, loss=0.3327, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2138/2855 [1:32:23<28:28,  2.38s/it, loss=0.2172, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2139/2855 [1:32:23<27:28,  2.30s/it, loss=0.2172, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2139/2855 [1:32:28<27:28,  2.30s/it, loss=0.0772, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2140/2855 [1:32:28<36:45,  3.09s/it, loss=0.0772, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2140/2855 [1:32:30<36:45,  3.09s/it, loss=0.0711, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2141/2855 [1:32:30<32:28,  2.73s/it, loss=0.0711, lr=2.94e-05]
Epoch 1/8:  75%|███████▍  | 2141/2855 [1:32:32<32:28,  2.73s/it, loss=0.3250, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2142/2855 [1:32:32<31:14,  2.63s/it, loss=0.3250, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2142/2855 [1:32:34<31:14,  2.63s/it, loss=0.1798, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2143/2855 [1:32:34<28:22,  2.39s/it, loss=0.1798, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2143/2855 [1:32:36<28:22,  2.39s/it, loss=0.1802, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2144/2855 [1:32:36<26:32,  2.24s/it, loss=0.1802, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2144/2855 [1:32:38<26:32,  2.24s/it, loss=0.0656, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2145/2855 [1:32:38<25:03,  2.12s/it, loss=0.0656, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2145/2855 [1:32:39<25:03,  2.12s/it, loss=0.1011, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2146/2855 [1:32:39<24:08,  2.04s/it, loss=0.1011, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2146/2855 [1:32:42<24:08,  2.04s/it, loss=0.1728, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2147/2855 [1:32:42<24:18,  2.06s/it, loss=0.1728, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2147/2855 [1:32:43<24:18,  2.06s/it, loss=0.1084, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2148/2855 [1:32:43<23:17,  1.98s/it, loss=0.1084, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2148/2855 [1:32:45<23:17,  1.98s/it, loss=0.1698, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2149/2855 [1:32:45<22:32,  1.92s/it, loss=0.1698, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2149/2855 [1:32:47<22:32,  1.92s/it, loss=0.3358, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2150/2855 [1:32:47<21:42,  1.85s/it, loss=0.3358, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2150/2855 [1:32:48<21:42,  1.85s/it, loss=0.1856, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2151/2855 [1:32:48<20:43,  1.77s/it, loss=0.1856, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2151/2855 [1:32:50<20:43,  1.77s/it, loss=0.1666, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2152/2855 [1:32:50<21:10,  1.81s/it, loss=0.1666, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2152/2855 [1:32:53<21:10,  1.81s/it, loss=0.0289, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2153/2855 [1:32:53<23:17,  1.99s/it, loss=0.0289, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2153/2855 [1:32:55<23:17,  1.99s/it, loss=0.1139, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2154/2855 [1:32:55<22:50,  1.95s/it, loss=0.1139, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2154/2855 [1:32:56<22:50,  1.95s/it, loss=0.2699, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2155/2855 [1:32:56<22:35,  1.94s/it, loss=0.2699, lr=2.94e-05]
Epoch 1/8:  75%|███████▌  | 2155/2855 [1:32:58<22:35,  1.94s/it, loss=0.2757, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2156/2855 [1:32:58<21:49,  1.87s/it, loss=0.2757, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2156/2855 [1:33:00<21:49,  1.87s/it, loss=0.2042, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2157/2855 [1:33:00<21:43,  1.87s/it, loss=0.2042, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2157/2855 [1:33:02<21:43,  1.87s/it, loss=0.1962, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2158/2855 [1:33:02<21:50,  1.88s/it, loss=0.1962, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2158/2855 [1:33:04<21:50,  1.88s/it, loss=0.0995, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2159/2855 [1:33:04<21:23,  1.84s/it, loss=0.0995, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2159/2855 [1:33:05<21:23,  1.84s/it, loss=0.2951, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2160/2855 [1:33:05<21:14,  1.83s/it, loss=0.2951, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2160/2855 [1:33:07<21:14,  1.83s/it, loss=0.1279, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2161/2855 [1:33:07<21:11,  1.83s/it, loss=0.1279, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2161/2855 [1:33:09<21:11,  1.83s/it, loss=0.2189, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2162/2855 [1:33:09<20:49,  1.80s/it, loss=0.2189, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2162/2855 [1:33:11<20:49,  1.80s/it, loss=0.2344, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2163/2855 [1:33:11<21:27,  1.86s/it, loss=0.2344, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2163/2855 [1:33:13<21:27,  1.86s/it, loss=0.0576, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2164/2855 [1:33:13<21:15,  1.85s/it, loss=0.0576, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2164/2855 [1:33:15<21:15,  1.85s/it, loss=0.1665, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2165/2855 [1:33:15<20:40,  1.80s/it, loss=0.1665, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2165/2855 [1:33:16<20:40,  1.80s/it, loss=0.3468, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2166/2855 [1:33:16<20:13,  1.76s/it, loss=0.3468, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2166/2855 [1:33:18<20:13,  1.76s/it, loss=0.2686, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2167/2855 [1:33:18<19:17,  1.68s/it, loss=0.2686, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2167/2855 [1:33:19<19:17,  1.68s/it, loss=0.1027, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2168/2855 [1:33:19<18:37,  1.63s/it, loss=0.1027, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2168/2855 [1:33:21<18:37,  1.63s/it, loss=0.0262, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2169/2855 [1:33:21<18:57,  1.66s/it, loss=0.0262, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2169/2855 [1:33:23<18:57,  1.66s/it, loss=0.4197, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2170/2855 [1:33:23<19:10,  1.68s/it, loss=0.4197, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2170/2855 [1:33:24<19:10,  1.68s/it, loss=0.2356, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2171/2855 [1:33:24<18:32,  1.63s/it, loss=0.2356, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2171/2855 [1:33:26<18:32,  1.63s/it, loss=0.2096, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2172/2855 [1:33:26<18:24,  1.62s/it, loss=0.2096, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2172/2855 [1:33:27<18:24,  1.62s/it, loss=0.1447, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2173/2855 [1:33:27<18:06,  1.59s/it, loss=0.1447, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2173/2855 [1:33:29<18:06,  1.59s/it, loss=0.2933, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2174/2855 [1:33:29<17:33,  1.55s/it, loss=0.2933, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2174/2855 [1:33:30<17:33,  1.55s/it, loss=0.0782, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2175/2855 [1:33:30<17:35,  1.55s/it, loss=0.0782, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2175/2855 [1:33:32<17:35,  1.55s/it, loss=0.2431, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2176/2855 [1:33:32<18:02,  1.59s/it, loss=0.2431, lr=2.94e-05]
Epoch 1/8:  76%|███████▌  | 2176/2855 [1:33:34<18:02,  1.59s/it, loss=0.2399, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2177/2855 [1:33:34<18:53,  1.67s/it, loss=0.2399, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2177/2855 [1:33:36<18:53,  1.67s/it, loss=0.2407, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2178/2855 [1:33:36<19:48,  1.76s/it, loss=0.2407, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2178/2855 [1:33:38<19:48,  1.76s/it, loss=0.1948, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2179/2855 [1:33:38<20:03,  1.78s/it, loss=0.1948, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2179/2855 [1:33:39<20:03,  1.78s/it, loss=0.2576, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2180/2855 [1:33:39<19:30,  1.73s/it, loss=0.2576, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2180/2855 [1:33:41<19:30,  1.73s/it, loss=0.1057, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2181/2855 [1:33:41<19:54,  1.77s/it, loss=0.1057, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2181/2855 [1:33:43<19:54,  1.77s/it, loss=0.2415, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2182/2855 [1:33:43<20:07,  1.79s/it, loss=0.2415, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2182/2855 [1:33:45<20:07,  1.79s/it, loss=0.1850, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2183/2855 [1:33:45<19:41,  1.76s/it, loss=0.1850, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2183/2855 [1:33:46<19:41,  1.76s/it, loss=0.1065, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2184/2855 [1:33:46<19:35,  1.75s/it, loss=0.1065, lr=2.94e-05]
Epoch 1/8:  76%|███████▋  | 2184/2855 [1:33:48<19:35,  1.75s/it, loss=0.0018, lr=2.94e-05]
Epoch 1/8:  77%|███████▋  | 2185/2855 [1:33:48<19:17,  1.73s/it, loss=0.0018, lr=2.94e-05]
Epoch 1/8:  77%|███████▋  | 2185/2855 [1:33:50<19:17,  1.73s/it, loss=0.1089, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2186/2855 [1:33:50<19:13,  1.72s/it, loss=0.1089, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2186/2855 [1:33:52<19:13,  1.72s/it, loss=0.2222, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2187/2855 [1:33:52<20:04,  1.80s/it, loss=0.2222, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2187/2855 [1:33:53<20:04,  1.80s/it, loss=0.1890, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2188/2855 [1:33:53<19:46,  1.78s/it, loss=0.1890, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2188/2855 [1:33:55<19:46,  1.78s/it, loss=0.1246, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2189/2855 [1:33:55<19:30,  1.76s/it, loss=0.1246, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2189/2855 [1:33:57<19:30,  1.76s/it, loss=0.1273, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2190/2855 [1:33:57<19:12,  1.73s/it, loss=0.1273, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2190/2855 [1:33:58<19:12,  1.73s/it, loss=0.1551, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2191/2855 [1:33:58<18:14,  1.65s/it, loss=0.1551, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2191/2855 [1:34:00<18:14,  1.65s/it, loss=0.1512, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2192/2855 [1:34:00<17:29,  1.58s/it, loss=0.1512, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2192/2855 [1:34:01<17:29,  1.58s/it, loss=0.1237, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2193/2855 [1:34:01<17:37,  1.60s/it, loss=0.1237, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2193/2855 [1:34:03<17:37,  1.60s/it, loss=0.1661, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2194/2855 [1:34:03<18:30,  1.68s/it, loss=0.1661, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2194/2855 [1:34:05<18:30,  1.68s/it, loss=0.0892, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2195/2855 [1:34:05<19:10,  1.74s/it, loss=0.0892, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2195/2855 [1:34:07<19:10,  1.74s/it, loss=0.1643, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2196/2855 [1:34:07<19:27,  1.77s/it, loss=0.1643, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2196/2855 [1:34:09<19:27,  1.77s/it, loss=0.0760, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2197/2855 [1:34:09<18:49,  1.72s/it, loss=0.0760, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2197/2855 [1:34:10<18:49,  1.72s/it, loss=0.0413, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2198/2855 [1:34:10<18:49,  1.72s/it, loss=0.0413, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2198/2855 [1:34:12<18:49,  1.72s/it, loss=0.2259, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2199/2855 [1:34:12<19:19,  1.77s/it, loss=0.2259, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2199/2855 [1:34:14<19:19,  1.77s/it, loss=0.1934, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2200/2855 [1:34:14<18:48,  1.72s/it, loss=0.1934, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2200/2855 [1:34:16<18:48,  1.72s/it, loss=0.0994, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2201/2855 [1:34:16<18:44,  1.72s/it, loss=0.0994, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2201/2855 [1:34:17<18:44,  1.72s/it, loss=0.1601, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2202/2855 [1:34:17<18:57,  1.74s/it, loss=0.1601, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2202/2855 [1:34:19<18:57,  1.74s/it, loss=0.1759, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2203/2855 [1:34:19<18:40,  1.72s/it, loss=0.1759, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2203/2855 [1:34:21<18:40,  1.72s/it, loss=0.2143, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2204/2855 [1:34:21<19:06,  1.76s/it, loss=0.2143, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2204/2855 [1:34:23<19:06,  1.76s/it, loss=0.1594, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2205/2855 [1:34:23<19:23,  1.79s/it, loss=0.1594, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2205/2855 [1:34:24<19:23,  1.79s/it, loss=0.2101, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2206/2855 [1:34:24<18:40,  1.73s/it, loss=0.2101, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2206/2855 [1:34:26<18:40,  1.73s/it, loss=0.1070, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2207/2855 [1:34:26<17:58,  1.66s/it, loss=0.1070, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2207/2855 [1:34:27<17:58,  1.66s/it, loss=0.1885, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2208/2855 [1:34:27<17:22,  1.61s/it, loss=0.1885, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2208/2855 [1:34:29<17:22,  1.61s/it, loss=0.1851, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2209/2855 [1:34:29<16:38,  1.55s/it, loss=0.1851, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2209/2855 [1:34:30<16:38,  1.55s/it, loss=0.2758, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2210/2855 [1:34:30<16:16,  1.51s/it, loss=0.2758, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2210/2855 [1:34:32<16:16,  1.51s/it, loss=0.2582, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2211/2855 [1:34:32<16:33,  1.54s/it, loss=0.2582, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2211/2855 [1:34:33<16:33,  1.54s/it, loss=0.0940, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2212/2855 [1:34:33<16:18,  1.52s/it, loss=0.0940, lr=2.93e-05]
Epoch 1/8:  77%|███████▋  | 2212/2855 [1:34:35<16:18,  1.52s/it, loss=0.3400, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2213/2855 [1:34:35<16:31,  1.54s/it, loss=0.3400, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2213/2855 [1:34:37<16:31,  1.54s/it, loss=0.0556, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2214/2855 [1:34:37<17:38,  1.65s/it, loss=0.0556, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2214/2855 [1:34:38<17:38,  1.65s/it, loss=0.1430, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2215/2855 [1:34:38<18:05,  1.70s/it, loss=0.1430, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2215/2855 [1:34:40<18:05,  1.70s/it, loss=0.2759, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2216/2855 [1:34:40<17:48,  1.67s/it, loss=0.2759, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2216/2855 [1:34:42<17:48,  1.67s/it, loss=0.3735, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2217/2855 [1:34:42<18:24,  1.73s/it, loss=0.3735, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2217/2855 [1:34:44<18:24,  1.73s/it, loss=0.1686, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2218/2855 [1:34:44<18:21,  1.73s/it, loss=0.1686, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2218/2855 [1:34:46<18:21,  1.73s/it, loss=0.2653, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2219/2855 [1:34:46<18:35,  1.75s/it, loss=0.2653, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2219/2855 [1:34:47<18:35,  1.75s/it, loss=0.6367, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2220/2855 [1:34:47<18:25,  1.74s/it, loss=0.6367, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2220/2855 [1:34:49<18:25,  1.74s/it, loss=0.2582, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2221/2855 [1:34:49<18:02,  1.71s/it, loss=0.2582, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2221/2855 [1:34:51<18:02,  1.71s/it, loss=0.1612, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2222/2855 [1:34:51<18:09,  1.72s/it, loss=0.1612, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2222/2855 [1:34:52<18:09,  1.72s/it, loss=0.0902, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2223/2855 [1:34:52<18:36,  1.77s/it, loss=0.0902, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2223/2855 [1:34:54<18:36,  1.77s/it, loss=0.2679, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2224/2855 [1:34:54<18:16,  1.74s/it, loss=0.2679, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2224/2855 [1:34:56<18:16,  1.74s/it, loss=0.1556, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2225/2855 [1:34:56<17:48,  1.70s/it, loss=0.1556, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2225/2855 [1:34:57<17:48,  1.70s/it, loss=0.2663, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2226/2855 [1:34:57<17:30,  1.67s/it, loss=0.2663, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2226/2855 [1:34:59<17:30,  1.67s/it, loss=0.2179, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2227/2855 [1:34:59<16:46,  1.60s/it, loss=0.2179, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2227/2855 [1:35:00<16:46,  1.60s/it, loss=0.3163, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2228/2855 [1:35:00<16:07,  1.54s/it, loss=0.3163, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2228/2855 [1:35:02<16:07,  1.54s/it, loss=0.3726, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2229/2855 [1:35:02<16:34,  1.59s/it, loss=0.3726, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2229/2855 [1:35:03<16:34,  1.59s/it, loss=0.1807, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2230/2855 [1:35:03<16:08,  1.55s/it, loss=0.1807, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2230/2855 [1:35:05<16:08,  1.55s/it, loss=0.2244, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2231/2855 [1:35:05<15:59,  1.54s/it, loss=0.2244, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2231/2855 [1:35:06<15:59,  1.54s/it, loss=0.3238, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2232/2855 [1:35:06<16:10,  1.56s/it, loss=0.3238, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2232/2855 [1:35:08<16:10,  1.56s/it, loss=0.4154, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2233/2855 [1:35:08<16:06,  1.55s/it, loss=0.4154, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2233/2855 [1:35:09<16:06,  1.55s/it, loss=0.2713, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2234/2855 [1:35:09<15:46,  1.52s/it, loss=0.2713, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2234/2855 [1:35:11<15:46,  1.52s/it, loss=0.2096, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2235/2855 [1:35:11<16:14,  1.57s/it, loss=0.2096, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2235/2855 [1:35:13<16:14,  1.57s/it, loss=0.2555, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2236/2855 [1:35:13<16:39,  1.61s/it, loss=0.2555, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2236/2855 [1:35:16<16:39,  1.61s/it, loss=0.1630, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2237/2855 [1:35:16<20:39,  2.01s/it, loss=0.1630, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2237/2855 [1:35:24<20:39,  2.01s/it, loss=0.1066, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2238/2855 [1:35:24<40:10,  3.91s/it, loss=0.1066, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2238/2855 [1:35:28<40:10,  3.91s/it, loss=0.2656, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2239/2855 [1:35:28<38:58,  3.80s/it, loss=0.2656, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2239/2855 [1:35:33<38:58,  3.80s/it, loss=0.1227, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2240/2855 [1:35:33<43:51,  4.28s/it, loss=0.1227, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2240/2855 [1:35:35<43:51,  4.28s/it, loss=0.2857, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2241/2855 [1:35:35<36:36,  3.58s/it, loss=0.2857, lr=2.93e-05]
Epoch 1/8:  78%|███████▊  | 2241/2855 [1:35:37<36:36,  3.58s/it, loss=0.2452, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2242/2855 [1:35:37<31:33,  3.09s/it, loss=0.2452, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2242/2855 [1:35:39<31:33,  3.09s/it, loss=0.2027, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2243/2855 [1:35:39<27:09,  2.66s/it, loss=0.2027, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2243/2855 [1:35:40<27:09,  2.66s/it, loss=0.2117, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2244/2855 [1:35:40<24:29,  2.40s/it, loss=0.2117, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2244/2855 [1:35:42<24:29,  2.40s/it, loss=0.1723, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2245/2855 [1:35:42<22:33,  2.22s/it, loss=0.1723, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2245/2855 [1:35:44<22:33,  2.22s/it, loss=0.3047, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2246/2855 [1:35:44<21:17,  2.10s/it, loss=0.3047, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2246/2855 [1:35:46<21:17,  2.10s/it, loss=0.0996, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2247/2855 [1:35:46<20:22,  2.01s/it, loss=0.0996, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2247/2855 [1:35:47<20:22,  2.01s/it, loss=0.1388, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2248/2855 [1:35:47<19:10,  1.90s/it, loss=0.1388, lr=2.93e-05]
Epoch 1/8:  79%|███████▊  | 2248/2855 [1:35:49<19:10,  1.90s/it, loss=0.3778, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2249/2855 [1:35:49<18:43,  1.85s/it, loss=0.3778, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2249/2855 [1:35:51<18:43,  1.85s/it, loss=0.2428, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2250/2855 [1:35:51<18:37,  1.85s/it, loss=0.2428, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2250/2855 [1:35:53<18:37,  1.85s/it, loss=0.1421, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2251/2855 [1:35:53<17:59,  1.79s/it, loss=0.1421, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2251/2855 [1:35:54<17:59,  1.79s/it, loss=0.2888, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2252/2855 [1:35:54<17:43,  1.76s/it, loss=0.2888, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2252/2855 [1:35:56<17:43,  1.76s/it, loss=0.2138, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2253/2855 [1:35:56<17:30,  1.74s/it, loss=0.2138, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2253/2855 [1:35:58<17:30,  1.74s/it, loss=0.2946, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2254/2855 [1:35:58<16:37,  1.66s/it, loss=0.2946, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2254/2855 [1:35:59<16:37,  1.66s/it, loss=0.1662, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2255/2855 [1:35:59<16:12,  1.62s/it, loss=0.1662, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2255/2855 [1:36:01<16:12,  1.62s/it, loss=0.1178, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2256/2855 [1:36:01<15:58,  1.60s/it, loss=0.1178, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2256/2855 [1:36:02<15:58,  1.60s/it, loss=0.0967, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2257/2855 [1:36:02<15:34,  1.56s/it, loss=0.0967, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2257/2855 [1:36:04<15:34,  1.56s/it, loss=0.3779, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2258/2855 [1:36:04<17:57,  1.80s/it, loss=0.3779, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2258/2855 [1:36:06<17:57,  1.80s/it, loss=0.1771, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2259/2855 [1:36:06<17:10,  1.73s/it, loss=0.1771, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2259/2855 [1:36:08<17:10,  1.73s/it, loss=0.1739, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2260/2855 [1:36:08<17:21,  1.75s/it, loss=0.1739, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2260/2855 [1:36:10<17:21,  1.75s/it, loss=0.2238, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2261/2855 [1:36:10<17:31,  1.77s/it, loss=0.2238, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2261/2855 [1:36:11<17:31,  1.77s/it, loss=0.0394, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2262/2855 [1:36:11<17:36,  1.78s/it, loss=0.0394, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2262/2855 [1:36:13<17:36,  1.78s/it, loss=0.1396, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2263/2855 [1:36:13<17:20,  1.76s/it, loss=0.1396, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2263/2855 [1:36:15<17:20,  1.76s/it, loss=0.0688, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2264/2855 [1:36:15<17:21,  1.76s/it, loss=0.0688, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2264/2855 [1:36:17<17:21,  1.76s/it, loss=0.1823, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2265/2855 [1:36:17<17:13,  1.75s/it, loss=0.1823, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2265/2855 [1:36:18<17:13,  1.75s/it, loss=0.1043, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2266/2855 [1:36:18<16:58,  1.73s/it, loss=0.1043, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2266/2855 [1:36:20<16:58,  1.73s/it, loss=0.3247, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2267/2855 [1:36:20<17:19,  1.77s/it, loss=0.3247, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2267/2855 [1:36:22<17:19,  1.77s/it, loss=0.2640, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2268/2855 [1:36:22<17:06,  1.75s/it, loss=0.2640, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2268/2855 [1:36:24<17:06,  1.75s/it, loss=0.1640, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2269/2855 [1:36:24<17:05,  1.75s/it, loss=0.1640, lr=2.93e-05]
Epoch 1/8:  79%|███████▉  | 2269/2855 [1:36:26<17:05,  1.75s/it, loss=0.1081, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2270/2855 [1:36:26<17:35,  1.80s/it, loss=0.1081, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2270/2855 [1:36:27<17:35,  1.80s/it, loss=0.4260, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2271/2855 [1:36:27<17:00,  1.75s/it, loss=0.4260, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2271/2855 [1:36:29<17:00,  1.75s/it, loss=0.0613, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2272/2855 [1:36:29<16:06,  1.66s/it, loss=0.0613, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2272/2855 [1:36:30<16:06,  1.66s/it, loss=0.1999, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2273/2855 [1:36:30<15:45,  1.62s/it, loss=0.1999, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2273/2855 [1:36:32<15:45,  1.62s/it, loss=0.1855, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2274/2855 [1:36:32<15:19,  1.58s/it, loss=0.1855, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2274/2855 [1:36:33<15:19,  1.58s/it, loss=0.2228, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2275/2855 [1:36:33<14:57,  1.55s/it, loss=0.2228, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2275/2855 [1:36:35<14:57,  1.55s/it, loss=0.2609, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2276/2855 [1:36:35<15:04,  1.56s/it, loss=0.2609, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2276/2855 [1:36:36<15:04,  1.56s/it, loss=0.5250, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2277/2855 [1:36:36<15:19,  1.59s/it, loss=0.5250, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2277/2855 [1:36:38<15:19,  1.59s/it, loss=0.2332, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2278/2855 [1:36:38<15:10,  1.58s/it, loss=0.2332, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2278/2855 [1:36:40<15:10,  1.58s/it, loss=0.1990, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2279/2855 [1:36:40<15:29,  1.61s/it, loss=0.1990, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2279/2855 [1:36:41<15:29,  1.61s/it, loss=0.0993, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2280/2855 [1:36:41<16:00,  1.67s/it, loss=0.0993, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2280/2855 [1:36:43<16:00,  1.67s/it, loss=0.1736, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2281/2855 [1:36:43<15:54,  1.66s/it, loss=0.1736, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2281/2855 [1:36:45<15:54,  1.66s/it, loss=0.4099, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2282/2855 [1:36:45<16:30,  1.73s/it, loss=0.4099, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2282/2855 [1:36:47<16:30,  1.73s/it, loss=0.0195, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2283/2855 [1:36:47<16:51,  1.77s/it, loss=0.0195, lr=2.93e-05]
Epoch 1/8:  80%|███████▉  | 2283/2855 [1:36:49<16:51,  1.77s/it, loss=0.1898, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2284/2855 [1:36:49<18:00,  1.89s/it, loss=0.1898, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2284/2855 [1:36:52<18:00,  1.89s/it, loss=0.0916, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2285/2855 [1:36:52<20:00,  2.11s/it, loss=0.0916, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2285/2855 [1:36:54<20:00,  2.11s/it, loss=0.0886, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2286/2855 [1:36:54<19:19,  2.04s/it, loss=0.0886, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2286/2855 [1:36:55<19:19,  2.04s/it, loss=0.1225, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2287/2855 [1:36:55<19:07,  2.02s/it, loss=0.1225, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2287/2855 [1:36:57<19:07,  2.02s/it, loss=0.2811, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2288/2855 [1:36:57<18:24,  1.95s/it, loss=0.2811, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2288/2855 [1:36:59<18:24,  1.95s/it, loss=0.3048, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2289/2855 [1:36:59<17:39,  1.87s/it, loss=0.3048, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2289/2855 [1:37:01<17:39,  1.87s/it, loss=0.1186, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2290/2855 [1:37:01<16:49,  1.79s/it, loss=0.1186, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2290/2855 [1:37:02<16:49,  1.79s/it, loss=0.1258, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2291/2855 [1:37:02<16:46,  1.79s/it, loss=0.1258, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2291/2855 [1:37:04<16:46,  1.79s/it, loss=0.1645, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2292/2855 [1:37:04<16:49,  1.79s/it, loss=0.1645, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2292/2855 [1:37:06<16:49,  1.79s/it, loss=0.1691, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2293/2855 [1:37:06<16:46,  1.79s/it, loss=0.1691, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2293/2855 [1:37:08<16:46,  1.79s/it, loss=0.2176, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2294/2855 [1:37:08<17:07,  1.83s/it, loss=0.2176, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2294/2855 [1:37:10<17:07,  1.83s/it, loss=0.2908, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2295/2855 [1:37:10<16:44,  1.79s/it, loss=0.2908, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2295/2855 [1:37:11<16:44,  1.79s/it, loss=0.2171, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2296/2855 [1:37:11<16:42,  1.79s/it, loss=0.2171, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2296/2855 [1:37:13<16:42,  1.79s/it, loss=0.1418, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2297/2855 [1:37:13<16:43,  1.80s/it, loss=0.1418, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2297/2855 [1:37:15<16:43,  1.80s/it, loss=0.1806, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2298/2855 [1:37:15<16:39,  1.79s/it, loss=0.1806, lr=2.93e-05]
Epoch 1/8:  80%|████████  | 2298/2855 [1:37:17<16:39,  1.79s/it, loss=0.1399, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2299/2855 [1:37:17<16:00,  1.73s/it, loss=0.1399, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2299/2855 [1:37:18<16:00,  1.73s/it, loss=0.1123, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2300/2855 [1:37:18<15:50,  1.71s/it, loss=0.1123, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2300/2855 [1:37:20<15:50,  1.71s/it, loss=0.2549, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2301/2855 [1:37:20<15:16,  1.65s/it, loss=0.2549, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2301/2855 [1:37:21<15:16,  1.65s/it, loss=0.0942, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2302/2855 [1:37:21<15:27,  1.68s/it, loss=0.0942, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2302/2855 [1:37:23<15:27,  1.68s/it, loss=0.2186, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2303/2855 [1:37:23<16:27,  1.79s/it, loss=0.2186, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2303/2855 [1:37:25<16:27,  1.79s/it, loss=0.1306, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2304/2855 [1:37:25<16:40,  1.82s/it, loss=0.1306, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2304/2855 [1:37:27<16:40,  1.82s/it, loss=0.2388, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2305/2855 [1:37:27<16:32,  1.80s/it, loss=0.2388, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2305/2855 [1:37:29<16:32,  1.80s/it, loss=0.1519, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2306/2855 [1:37:29<16:21,  1.79s/it, loss=0.1519, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2306/2855 [1:37:31<16:21,  1.79s/it, loss=0.1196, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2307/2855 [1:37:31<16:02,  1.76s/it, loss=0.1196, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2307/2855 [1:37:33<16:02,  1.76s/it, loss=0.3823, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2308/2855 [1:37:33<16:38,  1.83s/it, loss=0.3823, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2308/2855 [1:37:34<16:38,  1.83s/it, loss=0.1541, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2309/2855 [1:37:34<16:51,  1.85s/it, loss=0.1541, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2309/2855 [1:37:36<16:51,  1.85s/it, loss=0.0653, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2310/2855 [1:37:36<17:01,  1.87s/it, loss=0.0653, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2310/2855 [1:37:38<17:01,  1.87s/it, loss=0.1232, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2311/2855 [1:37:38<17:11,  1.90s/it, loss=0.1232, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2311/2855 [1:37:40<17:11,  1.90s/it, loss=0.4121, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2312/2855 [1:37:40<16:35,  1.83s/it, loss=0.4121, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2312/2855 [1:37:42<16:35,  1.83s/it, loss=0.2074, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2313/2855 [1:37:42<16:16,  1.80s/it, loss=0.2074, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2313/2855 [1:37:44<16:16,  1.80s/it, loss=0.0811, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2314/2855 [1:37:44<16:28,  1.83s/it, loss=0.0811, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2314/2855 [1:37:45<16:28,  1.83s/it, loss=0.3245, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2315/2855 [1:37:45<15:48,  1.76s/it, loss=0.3245, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2315/2855 [1:37:47<15:48,  1.76s/it, loss=0.1071, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2316/2855 [1:37:47<15:22,  1.71s/it, loss=0.1071, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2316/2855 [1:37:49<15:22,  1.71s/it, loss=0.2113, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2317/2855 [1:37:49<15:11,  1.69s/it, loss=0.2113, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2317/2855 [1:37:50<15:11,  1.69s/it, loss=0.4255, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2318/2855 [1:37:50<14:34,  1.63s/it, loss=0.4255, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2318/2855 [1:37:52<14:34,  1.63s/it, loss=0.1149, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2319/2855 [1:37:52<14:20,  1.61s/it, loss=0.1149, lr=2.93e-05]
Epoch 1/8:  81%|████████  | 2319/2855 [1:37:53<14:20,  1.61s/it, loss=0.0033, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2320/2855 [1:37:53<14:46,  1.66s/it, loss=0.0033, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2320/2855 [1:37:55<14:46,  1.66s/it, loss=0.1339, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2321/2855 [1:37:55<14:56,  1.68s/it, loss=0.1339, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2321/2855 [1:37:57<14:56,  1.68s/it, loss=0.2070, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2322/2855 [1:37:57<14:52,  1.68s/it, loss=0.2070, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2322/2855 [1:37:59<14:52,  1.68s/it, loss=0.0719, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2323/2855 [1:37:59<15:25,  1.74s/it, loss=0.0719, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2323/2855 [1:38:01<15:25,  1.74s/it, loss=0.1681, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2324/2855 [1:38:01<16:06,  1.82s/it, loss=0.1681, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2324/2855 [1:38:03<16:06,  1.82s/it, loss=0.2498, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2325/2855 [1:38:03<16:31,  1.87s/it, loss=0.2498, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2325/2855 [1:38:05<16:31,  1.87s/it, loss=0.1598, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2326/2855 [1:38:05<16:38,  1.89s/it, loss=0.1598, lr=2.93e-05]
Epoch 1/8:  81%|████████▏ | 2326/2855 [1:38:06<16:38,  1.89s/it, loss=0.2572, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2327/2855 [1:38:06<16:23,  1.86s/it, loss=0.2572, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2327/2855 [1:38:08<16:23,  1.86s/it, loss=0.1800, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2328/2855 [1:38:08<16:28,  1.88s/it, loss=0.1800, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2328/2855 [1:38:10<16:28,  1.88s/it, loss=0.3322, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2329/2855 [1:38:10<16:07,  1.84s/it, loss=0.3322, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2329/2855 [1:38:12<16:07,  1.84s/it, loss=0.3706, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2330/2855 [1:38:12<15:47,  1.80s/it, loss=0.3706, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2330/2855 [1:38:14<15:47,  1.80s/it, loss=0.1624, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2331/2855 [1:38:14<16:13,  1.86s/it, loss=0.1624, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2331/2855 [1:38:15<16:13,  1.86s/it, loss=0.2606, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2332/2855 [1:38:15<15:59,  1.84s/it, loss=0.2606, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2332/2855 [1:38:17<15:59,  1.84s/it, loss=0.1417, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2333/2855 [1:38:17<16:01,  1.84s/it, loss=0.1417, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2333/2855 [1:38:19<16:01,  1.84s/it, loss=0.2886, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2334/2855 [1:38:19<16:20,  1.88s/it, loss=0.2886, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2334/2855 [1:38:21<16:20,  1.88s/it, loss=0.1222, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2335/2855 [1:38:21<15:46,  1.82s/it, loss=0.1222, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2335/2855 [1:38:23<15:46,  1.82s/it, loss=0.1307, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2336/2855 [1:38:23<15:31,  1.80s/it, loss=0.1307, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2336/2855 [1:38:24<15:31,  1.80s/it, loss=0.0751, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2337/2855 [1:38:24<15:18,  1.77s/it, loss=0.0751, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2337/2855 [1:38:26<15:18,  1.77s/it, loss=0.3239, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2338/2855 [1:38:26<14:52,  1.73s/it, loss=0.3239, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2338/2855 [1:38:28<14:52,  1.73s/it, loss=0.2075, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2339/2855 [1:38:28<15:00,  1.75s/it, loss=0.2075, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2339/2855 [1:38:30<15:00,  1.75s/it, loss=0.3098, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2340/2855 [1:38:30<15:00,  1.75s/it, loss=0.3098, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2340/2855 [1:38:31<15:00,  1.75s/it, loss=0.1437, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2341/2855 [1:38:31<14:26,  1.69s/it, loss=0.1437, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2341/2855 [1:38:33<14:26,  1.69s/it, loss=0.2376, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2342/2855 [1:38:33<15:16,  1.79s/it, loss=0.2376, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2342/2855 [1:38:35<15:16,  1.79s/it, loss=0.2314, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2343/2855 [1:38:35<15:28,  1.81s/it, loss=0.2314, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2343/2855 [1:38:37<15:28,  1.81s/it, loss=0.2022, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2344/2855 [1:38:37<15:41,  1.84s/it, loss=0.2022, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2344/2855 [1:38:39<15:41,  1.84s/it, loss=0.1725, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2345/2855 [1:38:39<15:53,  1.87s/it, loss=0.1725, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2345/2855 [1:38:41<15:53,  1.87s/it, loss=0.2028, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2346/2855 [1:38:41<15:27,  1.82s/it, loss=0.2028, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2346/2855 [1:38:42<15:27,  1.82s/it, loss=0.0683, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2347/2855 [1:38:42<15:15,  1.80s/it, loss=0.0683, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2347/2855 [1:38:44<15:15,  1.80s/it, loss=0.1684, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2348/2855 [1:38:44<15:53,  1.88s/it, loss=0.1684, lr=2.93e-05]
Epoch 1/8:  82%|████████▏ | 2348/2855 [1:38:46<15:53,  1.88s/it, loss=0.1383, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2349/2855 [1:38:46<15:34,  1.85s/it, loss=0.1383, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2349/2855 [1:38:48<15:34,  1.85s/it, loss=0.1208, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2350/2855 [1:38:48<15:36,  1.86s/it, loss=0.1208, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2350/2855 [1:38:50<15:36,  1.86s/it, loss=0.2679, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2351/2855 [1:38:50<15:35,  1.86s/it, loss=0.2679, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2351/2855 [1:38:52<15:35,  1.86s/it, loss=0.1845, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2352/2855 [1:38:52<15:13,  1.82s/it, loss=0.1845, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2352/2855 [1:38:54<15:13,  1.82s/it, loss=0.1047, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2353/2855 [1:38:54<15:37,  1.87s/it, loss=0.1047, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2353/2855 [1:38:55<15:37,  1.87s/it, loss=0.3083, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2354/2855 [1:38:55<15:16,  1.83s/it, loss=0.3083, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2354/2855 [1:38:57<15:16,  1.83s/it, loss=0.1493, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2355/2855 [1:38:57<14:46,  1.77s/it, loss=0.1493, lr=2.92e-05]
Epoch 1/8:  82%|████████▏ | 2355/2855 [1:38:59<14:46,  1.77s/it, loss=0.1072, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2356/2855 [1:38:59<14:45,  1.77s/it, loss=0.1072, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2356/2855 [1:39:00<14:45,  1.77s/it, loss=0.0779, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2357/2855 [1:39:00<14:00,  1.69s/it, loss=0.0779, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2357/2855 [1:39:02<14:00,  1.69s/it, loss=0.3235, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2358/2855 [1:39:02<14:04,  1.70s/it, loss=0.3235, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2358/2855 [1:39:04<14:04,  1.70s/it, loss=0.2214, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2359/2855 [1:39:04<15:25,  1.87s/it, loss=0.2214, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2359/2855 [1:39:06<15:25,  1.87s/it, loss=0.3109, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2360/2855 [1:39:06<15:20,  1.86s/it, loss=0.3109, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2360/2855 [1:39:08<15:20,  1.86s/it, loss=0.2541, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2361/2855 [1:39:08<15:14,  1.85s/it, loss=0.2541, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2361/2855 [1:39:10<15:14,  1.85s/it, loss=0.1679, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2362/2855 [1:39:10<15:16,  1.86s/it, loss=0.1679, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2362/2855 [1:39:12<15:16,  1.86s/it, loss=0.0373, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2363/2855 [1:39:12<14:56,  1.82s/it, loss=0.0373, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2363/2855 [1:39:13<14:56,  1.82s/it, loss=0.1554, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2364/2855 [1:39:13<15:01,  1.84s/it, loss=0.1554, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2364/2855 [1:39:15<15:01,  1.84s/it, loss=0.1832, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2365/2855 [1:39:15<15:03,  1.84s/it, loss=0.1832, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2365/2855 [1:39:17<15:03,  1.84s/it, loss=0.2223, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2366/2855 [1:39:17<14:52,  1.82s/it, loss=0.2223, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2366/2855 [1:39:19<14:52,  1.82s/it, loss=0.3094, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2367/2855 [1:39:19<15:16,  1.88s/it, loss=0.3094, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2367/2855 [1:39:21<15:16,  1.88s/it, loss=0.0966, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2368/2855 [1:39:21<14:55,  1.84s/it, loss=0.0966, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2368/2855 [1:39:23<14:55,  1.84s/it, loss=0.1296, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2369/2855 [1:39:23<14:50,  1.83s/it, loss=0.1296, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2369/2855 [1:39:25<14:50,  1.83s/it, loss=0.1267, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2370/2855 [1:39:25<15:28,  1.91s/it, loss=0.1267, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2370/2855 [1:39:26<15:28,  1.91s/it, loss=0.1154, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2371/2855 [1:39:26<14:52,  1.84s/it, loss=0.1154, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2371/2855 [1:39:28<14:52,  1.84s/it, loss=0.2282, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2372/2855 [1:39:28<14:38,  1.82s/it, loss=0.2282, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2372/2855 [1:39:30<14:38,  1.82s/it, loss=0.2305, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2373/2855 [1:39:30<14:19,  1.78s/it, loss=0.2305, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2373/2855 [1:39:31<14:19,  1.78s/it, loss=0.1093, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2374/2855 [1:39:31<13:45,  1.72s/it, loss=0.1093, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2374/2855 [1:39:33<13:45,  1.72s/it, loss=0.2464, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2375/2855 [1:39:33<13:34,  1.70s/it, loss=0.2464, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2375/2855 [1:39:35<13:34,  1.70s/it, loss=0.4142, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2376/2855 [1:39:35<13:30,  1.69s/it, loss=0.4142, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2376/2855 [1:39:37<13:30,  1.69s/it, loss=0.2449, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2377/2855 [1:39:37<14:25,  1.81s/it, loss=0.2449, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2377/2855 [1:39:39<14:25,  1.81s/it, loss=0.1643, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2378/2855 [1:39:39<15:18,  1.92s/it, loss=0.1643, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2378/2855 [1:39:41<15:18,  1.92s/it, loss=0.1806, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2379/2855 [1:39:41<14:51,  1.87s/it, loss=0.1806, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2379/2855 [1:39:43<14:51,  1.87s/it, loss=0.1693, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2380/2855 [1:39:43<14:33,  1.84s/it, loss=0.1693, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2380/2855 [1:39:45<14:33,  1.84s/it, loss=0.1082, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2381/2855 [1:39:45<14:52,  1.88s/it, loss=0.1082, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2381/2855 [1:39:46<14:52,  1.88s/it, loss=0.4065, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2382/2855 [1:39:46<14:37,  1.86s/it, loss=0.4065, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2382/2855 [1:39:48<14:37,  1.86s/it, loss=0.2187, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2383/2855 [1:39:48<14:38,  1.86s/it, loss=0.2187, lr=2.92e-05]
Epoch 1/8:  83%|████████▎ | 2383/2855 [1:39:50<14:38,  1.86s/it, loss=0.1357, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2384/2855 [1:39:50<14:50,  1.89s/it, loss=0.1357, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2384/2855 [1:39:52<14:50,  1.89s/it, loss=0.1377, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2385/2855 [1:39:52<14:22,  1.83s/it, loss=0.1377, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2385/2855 [1:39:54<14:22,  1.83s/it, loss=0.1814, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2386/2855 [1:39:54<14:38,  1.87s/it, loss=0.1814, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2386/2855 [1:39:56<14:38,  1.87s/it, loss=0.1652, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2387/2855 [1:39:56<14:55,  1.91s/it, loss=0.1652, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2387/2855 [1:39:58<14:55,  1.91s/it, loss=0.1139, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2388/2855 [1:39:58<14:36,  1.88s/it, loss=0.1139, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2388/2855 [1:40:00<14:36,  1.88s/it, loss=0.2260, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2389/2855 [1:40:00<14:34,  1.88s/it, loss=0.2260, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2389/2855 [1:40:01<14:34,  1.88s/it, loss=0.1518, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2390/2855 [1:40:01<13:38,  1.76s/it, loss=0.1518, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2390/2855 [1:40:02<13:38,  1.76s/it, loss=0.1088, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2391/2855 [1:40:02<13:00,  1.68s/it, loss=0.1088, lr=2.92e-05]
Epoch 1/8:  84%|████████▎ | 2391/2855 [1:40:05<13:00,  1.68s/it, loss=0.1781, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2392/2855 [1:40:05<14:05,  1.83s/it, loss=0.1781, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2392/2855 [1:40:07<14:05,  1.83s/it, loss=0.1616, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2393/2855 [1:40:07<14:19,  1.86s/it, loss=0.1616, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2393/2855 [1:40:09<14:19,  1.86s/it, loss=0.1366, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2394/2855 [1:40:09<14:30,  1.89s/it, loss=0.1366, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2394/2855 [1:40:10<14:30,  1.89s/it, loss=0.2241, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2395/2855 [1:40:10<14:29,  1.89s/it, loss=0.2241, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2395/2855 [1:40:12<14:29,  1.89s/it, loss=0.0709, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2396/2855 [1:40:12<14:05,  1.84s/it, loss=0.0709, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2396/2855 [1:40:14<14:05,  1.84s/it, loss=0.2087, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2397/2855 [1:40:14<14:07,  1.85s/it, loss=0.2087, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2397/2855 [1:40:16<14:07,  1.85s/it, loss=0.1808, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2398/2855 [1:40:16<14:16,  1.87s/it, loss=0.1808, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2398/2855 [1:40:18<14:16,  1.87s/it, loss=0.2961, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2399/2855 [1:40:18<14:04,  1.85s/it, loss=0.2961, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2399/2855 [1:40:20<14:04,  1.85s/it, loss=0.1810, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2400/2855 [1:40:20<14:31,  1.91s/it, loss=0.1810, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2400/2855 [1:40:22<14:31,  1.91s/it, loss=0.0734, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2401/2855 [1:40:22<14:13,  1.88s/it, loss=0.0734, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2401/2855 [1:40:23<14:13,  1.88s/it, loss=0.3726, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2402/2855 [1:40:23<14:00,  1.85s/it, loss=0.3726, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2402/2855 [1:40:25<14:00,  1.85s/it, loss=0.1662, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2403/2855 [1:40:25<14:21,  1.91s/it, loss=0.1662, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2403/2855 [1:40:27<14:21,  1.91s/it, loss=0.0991, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2404/2855 [1:40:27<14:00,  1.86s/it, loss=0.0991, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2404/2855 [1:40:29<14:00,  1.86s/it, loss=0.4198, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2405/2855 [1:40:29<13:42,  1.83s/it, loss=0.4198, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2405/2855 [1:40:31<13:42,  1.83s/it, loss=0.3897, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2406/2855 [1:40:31<13:37,  1.82s/it, loss=0.3897, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2406/2855 [1:40:32<13:37,  1.82s/it, loss=0.1376, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2407/2855 [1:40:32<13:03,  1.75s/it, loss=0.1376, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2407/2855 [1:40:34<13:03,  1.75s/it, loss=0.2024, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2408/2855 [1:40:34<13:14,  1.78s/it, loss=0.2024, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2408/2855 [1:40:36<13:14,  1.78s/it, loss=0.1150, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2409/2855 [1:40:36<13:22,  1.80s/it, loss=0.1150, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2409/2855 [1:40:38<13:22,  1.80s/it, loss=0.3152, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2410/2855 [1:40:38<12:56,  1.75s/it, loss=0.3152, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2410/2855 [1:40:40<12:56,  1.75s/it, loss=0.3197, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2411/2855 [1:40:40<13:35,  1.84s/it, loss=0.3197, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2411/2855 [1:40:42<13:35,  1.84s/it, loss=0.0921, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2412/2855 [1:40:42<13:52,  1.88s/it, loss=0.0921, lr=2.92e-05]
Epoch 1/8:  84%|████████▍ | 2412/2855 [1:40:44<13:52,  1.88s/it, loss=0.0915, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2413/2855 [1:40:44<13:53,  1.89s/it, loss=0.0915, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2413/2855 [1:40:46<13:53,  1.89s/it, loss=0.1145, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2414/2855 [1:40:46<13:59,  1.90s/it, loss=0.1145, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2414/2855 [1:40:47<13:59,  1.90s/it, loss=0.0381, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2415/2855 [1:40:47<13:41,  1.87s/it, loss=0.0381, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2415/2855 [1:40:49<13:41,  1.87s/it, loss=0.0665, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2416/2855 [1:40:49<13:53,  1.90s/it, loss=0.0665, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2416/2855 [1:40:51<13:53,  1.90s/it, loss=0.2643, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2417/2855 [1:40:51<14:03,  1.93s/it, loss=0.2643, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2417/2855 [1:40:53<14:03,  1.93s/it, loss=0.2532, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2418/2855 [1:40:53<13:37,  1.87s/it, loss=0.2532, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2418/2855 [1:40:55<13:37,  1.87s/it, loss=0.0761, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2419/2855 [1:40:55<14:07,  1.94s/it, loss=0.0761, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2419/2855 [1:40:57<14:07,  1.94s/it, loss=0.3144, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2420/2855 [1:40:57<13:56,  1.92s/it, loss=0.3144, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2420/2855 [1:40:59<13:56,  1.92s/it, loss=0.2322, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2421/2855 [1:40:59<13:42,  1.90s/it, loss=0.2322, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2421/2855 [1:41:01<13:42,  1.90s/it, loss=0.0053, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2422/2855 [1:41:01<14:00,  1.94s/it, loss=0.0053, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2422/2855 [1:41:03<14:00,  1.94s/it, loss=0.2353, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2423/2855 [1:41:03<13:38,  1.89s/it, loss=0.2353, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2423/2855 [1:41:04<13:38,  1.89s/it, loss=0.1368, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2424/2855 [1:41:04<13:17,  1.85s/it, loss=0.1368, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2424/2855 [1:41:06<13:17,  1.85s/it, loss=0.0383, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2425/2855 [1:41:06<12:53,  1.80s/it, loss=0.0383, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2425/2855 [1:41:08<12:53,  1.80s/it, loss=0.0551, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2426/2855 [1:41:08<12:26,  1.74s/it, loss=0.0551, lr=2.92e-05]
Epoch 1/8:  85%|████████▍ | 2426/2855 [1:41:10<12:26,  1.74s/it, loss=0.3814, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2427/2855 [1:41:10<13:05,  1.84s/it, loss=0.3814, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2427/2855 [1:41:12<13:05,  1.84s/it, loss=0.2422, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2428/2855 [1:41:12<13:39,  1.92s/it, loss=0.2422, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2428/2855 [1:41:14<13:39,  1.92s/it, loss=0.2741, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2429/2855 [1:41:14<13:34,  1.91s/it, loss=0.2741, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2429/2855 [1:41:16<13:34,  1.91s/it, loss=0.1026, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2430/2855 [1:41:16<13:38,  1.92s/it, loss=0.1026, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2430/2855 [1:41:18<13:38,  1.92s/it, loss=0.1950, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2431/2855 [1:41:18<13:25,  1.90s/it, loss=0.1950, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2431/2855 [1:41:20<13:25,  1.90s/it, loss=0.2136, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2432/2855 [1:41:20<13:28,  1.91s/it, loss=0.2136, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2432/2855 [1:41:21<13:28,  1.91s/it, loss=0.1430, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2433/2855 [1:41:21<13:25,  1.91s/it, loss=0.1430, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2433/2855 [1:41:23<13:25,  1.91s/it, loss=0.0342, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2434/2855 [1:41:23<13:08,  1.87s/it, loss=0.0342, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2434/2855 [1:41:25<13:08,  1.87s/it, loss=0.3781, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2435/2855 [1:41:25<13:41,  1.96s/it, loss=0.3781, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2435/2855 [1:41:27<13:41,  1.96s/it, loss=0.1666, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2436/2855 [1:41:27<13:35,  1.95s/it, loss=0.1666, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2436/2855 [1:41:29<13:35,  1.95s/it, loss=0.1222, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2437/2855 [1:41:29<13:19,  1.91s/it, loss=0.1222, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2437/2855 [1:41:31<13:19,  1.91s/it, loss=0.2988, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2438/2855 [1:41:31<13:32,  1.95s/it, loss=0.2988, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2438/2855 [1:41:33<13:32,  1.95s/it, loss=0.2071, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2439/2855 [1:41:33<12:59,  1.87s/it, loss=0.2071, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2439/2855 [1:41:35<12:59,  1.87s/it, loss=0.2902, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2440/2855 [1:41:35<12:42,  1.84s/it, loss=0.2902, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2440/2855 [1:41:36<12:42,  1.84s/it, loss=0.1033, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2441/2855 [1:41:36<12:33,  1.82s/it, loss=0.1033, lr=2.92e-05]
Epoch 1/8:  85%|████████▌ | 2441/2855 [1:41:38<12:33,  1.82s/it, loss=0.2651, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2442/2855 [1:41:38<12:18,  1.79s/it, loss=0.2651, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2442/2855 [1:41:40<12:18,  1.79s/it, loss=0.1934, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2443/2855 [1:41:40<12:42,  1.85s/it, loss=0.1934, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2443/2855 [1:41:42<12:42,  1.85s/it, loss=0.2847, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2444/2855 [1:41:42<12:19,  1.80s/it, loss=0.2847, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2444/2855 [1:41:43<12:19,  1.80s/it, loss=0.1850, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2445/2855 [1:41:43<11:47,  1.72s/it, loss=0.1850, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2445/2855 [1:41:45<11:47,  1.72s/it, loss=0.0892, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2446/2855 [1:41:45<12:07,  1.78s/it, loss=0.0892, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2446/2855 [1:41:47<12:07,  1.78s/it, loss=0.2785, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2447/2855 [1:41:47<11:55,  1.75s/it, loss=0.2785, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2447/2855 [1:41:49<11:55,  1.75s/it, loss=0.2485, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2448/2855 [1:41:49<11:38,  1.72s/it, loss=0.2485, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2448/2855 [1:41:51<11:38,  1.72s/it, loss=0.1327, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2449/2855 [1:41:51<12:16,  1.81s/it, loss=0.1327, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2449/2855 [1:41:52<12:16,  1.81s/it, loss=0.2081, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2450/2855 [1:41:52<11:52,  1.76s/it, loss=0.2081, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2450/2855 [1:41:54<11:52,  1.76s/it, loss=0.2089, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2451/2855 [1:41:54<12:23,  1.84s/it, loss=0.2089, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2451/2855 [1:41:56<12:23,  1.84s/it, loss=0.2816, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2452/2855 [1:41:56<13:00,  1.94s/it, loss=0.2816, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2452/2855 [1:41:58<13:00,  1.94s/it, loss=0.2877, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2453/2855 [1:41:58<12:38,  1.89s/it, loss=0.2877, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2453/2855 [1:42:00<12:38,  1.89s/it, loss=0.0994, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2454/2855 [1:42:00<12:57,  1.94s/it, loss=0.0994, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2454/2855 [1:42:02<12:57,  1.94s/it, loss=0.3098, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2455/2855 [1:42:02<12:48,  1.92s/it, loss=0.3098, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2455/2855 [1:42:04<12:48,  1.92s/it, loss=0.2365, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2456/2855 [1:42:04<12:24,  1.86s/it, loss=0.2365, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2456/2855 [1:42:06<12:24,  1.86s/it, loss=0.1228, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2457/2855 [1:42:06<12:45,  1.92s/it, loss=0.1228, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2457/2855 [1:42:08<12:45,  1.92s/it, loss=0.1713, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2458/2855 [1:42:08<12:44,  1.93s/it, loss=0.1713, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2458/2855 [1:42:10<12:44,  1.93s/it, loss=0.2067, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2459/2855 [1:42:10<12:51,  1.95s/it, loss=0.2067, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2459/2855 [1:42:12<12:51,  1.95s/it, loss=0.2623, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2460/2855 [1:42:12<12:46,  1.94s/it, loss=0.2623, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2460/2855 [1:42:14<12:46,  1.94s/it, loss=0.1375, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2461/2855 [1:42:14<12:23,  1.89s/it, loss=0.1375, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2461/2855 [1:42:15<12:23,  1.89s/it, loss=0.2467, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2462/2855 [1:42:15<12:29,  1.91s/it, loss=0.2467, lr=2.92e-05]
Epoch 1/8:  86%|████████▌ | 2462/2855 [1:42:17<12:29,  1.91s/it, loss=0.2565, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2463/2855 [1:42:17<12:31,  1.92s/it, loss=0.2565, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2463/2855 [1:42:19<12:31,  1.92s/it, loss=0.1146, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2464/2855 [1:42:19<12:05,  1.86s/it, loss=0.1146, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2464/2855 [1:42:21<12:05,  1.86s/it, loss=0.1505, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2465/2855 [1:42:21<11:57,  1.84s/it, loss=0.1505, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2465/2855 [1:42:22<11:57,  1.84s/it, loss=0.0795, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2466/2855 [1:42:22<11:21,  1.75s/it, loss=0.0795, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2466/2855 [1:42:24<11:21,  1.75s/it, loss=0.2221, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2467/2855 [1:42:24<11:10,  1.73s/it, loss=0.2221, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2467/2855 [1:42:26<11:10,  1.73s/it, loss=0.0931, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2468/2855 [1:42:26<11:51,  1.84s/it, loss=0.0931, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2468/2855 [1:42:28<11:51,  1.84s/it, loss=0.1720, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2469/2855 [1:42:28<11:21,  1.76s/it, loss=0.1720, lr=2.92e-05]
Epoch 1/8:  86%|████████▋ | 2469/2855 [1:42:30<11:21,  1.76s/it, loss=0.1857, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2470/2855 [1:42:30<11:14,  1.75s/it, loss=0.1857, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2470/2855 [1:42:32<11:14,  1.75s/it, loss=0.1266, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2471/2855 [1:42:32<11:39,  1.82s/it, loss=0.1266, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2471/2855 [1:42:34<11:39,  1.82s/it, loss=0.4078, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2472/2855 [1:42:34<12:03,  1.89s/it, loss=0.4078, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2472/2855 [1:42:36<12:03,  1.89s/it, loss=0.3367, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2473/2855 [1:42:36<12:33,  1.97s/it, loss=0.3367, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2473/2855 [1:42:38<12:33,  1.97s/it, loss=0.0947, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2474/2855 [1:42:38<12:34,  1.98s/it, loss=0.0947, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2474/2855 [1:42:40<12:34,  1.98s/it, loss=0.0773, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2475/2855 [1:42:40<12:21,  1.95s/it, loss=0.0773, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2475/2855 [1:42:42<12:21,  1.95s/it, loss=0.1753, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2476/2855 [1:42:42<12:13,  1.94s/it, loss=0.1753, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2476/2855 [1:42:43<12:13,  1.94s/it, loss=0.2597, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2477/2855 [1:42:43<11:57,  1.90s/it, loss=0.2597, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2477/2855 [1:42:45<11:57,  1.90s/it, loss=0.2190, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2478/2855 [1:42:45<11:50,  1.88s/it, loss=0.2190, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2478/2855 [1:42:47<11:50,  1.88s/it, loss=0.0020, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2479/2855 [1:42:47<12:12,  1.95s/it, loss=0.0020, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2479/2855 [1:42:49<12:12,  1.95s/it, loss=0.0872, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2480/2855 [1:42:49<11:58,  1.92s/it, loss=0.0872, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2480/2855 [1:42:51<11:58,  1.92s/it, loss=0.3578, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2481/2855 [1:42:51<12:04,  1.94s/it, loss=0.3578, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2481/2855 [1:42:53<12:04,  1.94s/it, loss=0.1189, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2482/2855 [1:42:53<11:50,  1.90s/it, loss=0.1189, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2482/2855 [1:42:55<11:50,  1.90s/it, loss=0.1479, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2483/2855 [1:42:55<11:41,  1.89s/it, loss=0.1479, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2483/2855 [1:42:57<11:41,  1.89s/it, loss=0.2224, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2484/2855 [1:42:57<11:46,  1.90s/it, loss=0.2224, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2484/2855 [1:42:58<11:46,  1.90s/it, loss=0.1726, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2485/2855 [1:42:58<11:02,  1.79s/it, loss=0.1726, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2485/2855 [1:43:00<11:02,  1.79s/it, loss=0.0039, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2486/2855 [1:43:00<11:18,  1.84s/it, loss=0.0039, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2486/2855 [1:43:02<11:18,  1.84s/it, loss=0.0852, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2487/2855 [1:43:02<11:57,  1.95s/it, loss=0.0852, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2487/2855 [1:43:04<11:57,  1.95s/it, loss=0.3606, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2488/2855 [1:43:04<11:32,  1.89s/it, loss=0.3606, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2488/2855 [1:43:06<11:32,  1.89s/it, loss=0.1112, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2489/2855 [1:43:06<11:43,  1.92s/it, loss=0.1112, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2489/2855 [1:43:08<11:43,  1.92s/it, loss=0.1633, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2490/2855 [1:43:08<11:38,  1.91s/it, loss=0.1633, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2490/2855 [1:43:10<11:38,  1.91s/it, loss=0.2059, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2491/2855 [1:43:10<11:31,  1.90s/it, loss=0.2059, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2491/2855 [1:43:12<11:31,  1.90s/it, loss=0.1875, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2492/2855 [1:43:12<11:53,  1.97s/it, loss=0.1875, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2492/2855 [1:43:14<11:53,  1.97s/it, loss=0.2875, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2493/2855 [1:43:14<11:41,  1.94s/it, loss=0.2875, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2493/2855 [1:43:16<11:41,  1.94s/it, loss=0.1722, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2494/2855 [1:43:16<11:31,  1.91s/it, loss=0.1722, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2494/2855 [1:43:18<11:31,  1.91s/it, loss=0.3156, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2495/2855 [1:43:18<11:42,  1.95s/it, loss=0.3156, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2495/2855 [1:43:20<11:42,  1.95s/it, loss=0.0806, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2496/2855 [1:43:20<11:40,  1.95s/it, loss=0.0806, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2496/2855 [1:43:22<11:40,  1.95s/it, loss=0.0342, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2497/2855 [1:43:22<11:47,  1.98s/it, loss=0.0342, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2497/2855 [1:43:24<11:47,  1.98s/it, loss=0.1725, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2498/2855 [1:43:24<11:28,  1.93s/it, loss=0.1725, lr=2.92e-05]
Epoch 1/8:  87%|████████▋ | 2498/2855 [1:43:25<11:28,  1.93s/it, loss=0.1862, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2499/2855 [1:43:25<11:03,  1.86s/it, loss=0.1862, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2499/2855 [1:43:27<11:03,  1.86s/it, loss=0.3551, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2500/2855 [1:43:27<11:01,  1.86s/it, loss=0.3551, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2500/2855 [1:43:29<11:01,  1.86s/it, loss=0.2157, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2501/2855 [1:43:29<10:35,  1.80s/it, loss=0.2157, lr=2.92e-05]
Epoch 1/8:  88%|████████▊ | 2501/2855 [1:43:31<10:35,  1.80s/it, loss=0.1312, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2502/2855 [1:43:31<10:41,  1.82s/it, loss=0.1312, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2502/2855 [1:43:32<10:41,  1.82s/it, loss=0.0956, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2503/2855 [1:43:32<10:23,  1.77s/it, loss=0.0956, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2503/2855 [1:43:34<10:23,  1.77s/it, loss=0.2587, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2504/2855 [1:43:34<09:59,  1.71s/it, loss=0.2587, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2504/2855 [1:43:36<09:59,  1.71s/it, loss=0.2384, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2505/2855 [1:43:36<10:19,  1.77s/it, loss=0.2384, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2505/2855 [1:43:38<10:19,  1.77s/it, loss=0.2020, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2506/2855 [1:43:38<10:25,  1.79s/it, loss=0.2020, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2506/2855 [1:43:40<10:25,  1.79s/it, loss=0.0898, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2507/2855 [1:43:40<11:01,  1.90s/it, loss=0.0898, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2507/2855 [1:43:42<11:01,  1.90s/it, loss=0.2562, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2508/2855 [1:43:42<11:29,  1.99s/it, loss=0.2562, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2508/2855 [1:43:44<11:29,  1.99s/it, loss=0.0446, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2509/2855 [1:43:44<11:12,  1.94s/it, loss=0.0446, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2509/2855 [1:43:46<11:12,  1.94s/it, loss=0.0667, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2510/2855 [1:43:46<10:55,  1.90s/it, loss=0.0667, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2510/2855 [1:43:48<10:55,  1.90s/it, loss=0.2402, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2511/2855 [1:43:48<11:06,  1.94s/it, loss=0.2402, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2511/2855 [1:43:50<11:06,  1.94s/it, loss=0.2159, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2512/2855 [1:43:50<11:02,  1.93s/it, loss=0.2159, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2512/2855 [1:43:52<11:02,  1.93s/it, loss=0.1381, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2513/2855 [1:43:52<11:09,  1.96s/it, loss=0.1381, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2513/2855 [1:43:54<11:09,  1.96s/it, loss=0.0341, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2514/2855 [1:43:54<10:59,  1.93s/it, loss=0.0341, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2514/2855 [1:43:55<10:59,  1.93s/it, loss=0.2417, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2515/2855 [1:43:55<10:50,  1.91s/it, loss=0.2417, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2515/2855 [1:43:57<10:50,  1.91s/it, loss=0.1381, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2516/2855 [1:43:57<11:04,  1.96s/it, loss=0.1381, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2516/2855 [1:43:59<11:04,  1.96s/it, loss=0.1868, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2517/2855 [1:43:59<10:54,  1.94s/it, loss=0.1868, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2517/2855 [1:44:01<10:54,  1.94s/it, loss=0.2284, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2518/2855 [1:44:01<11:02,  1.97s/it, loss=0.2284, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2518/2855 [1:44:03<11:02,  1.97s/it, loss=0.1212, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2519/2855 [1:44:03<10:54,  1.95s/it, loss=0.1212, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2519/2855 [1:44:05<10:54,  1.95s/it, loss=0.1295, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2520/2855 [1:44:05<10:09,  1.82s/it, loss=0.1295, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2520/2855 [1:44:07<10:09,  1.82s/it, loss=0.1898, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2521/2855 [1:44:07<10:32,  1.89s/it, loss=0.1898, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2521/2855 [1:44:09<10:32,  1.89s/it, loss=0.0953, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2522/2855 [1:44:09<10:49,  1.95s/it, loss=0.0953, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2522/2855 [1:44:11<10:49,  1.95s/it, loss=0.1960, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2523/2855 [1:44:11<10:46,  1.95s/it, loss=0.1960, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2523/2855 [1:44:13<10:46,  1.95s/it, loss=0.3882, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2524/2855 [1:44:13<10:55,  1.98s/it, loss=0.3882, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2524/2855 [1:44:15<10:55,  1.98s/it, loss=0.0021, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2525/2855 [1:44:15<10:39,  1.94s/it, loss=0.0021, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2525/2855 [1:44:17<10:39,  1.94s/it, loss=0.1262, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2526/2855 [1:44:17<10:37,  1.94s/it, loss=0.1262, lr=2.91e-05]
Epoch 1/8:  88%|████████▊ | 2526/2855 [1:44:19<10:37,  1.94s/it, loss=0.2268, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2527/2855 [1:44:19<10:43,  1.96s/it, loss=0.2268, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2527/2855 [1:44:21<10:43,  1.96s/it, loss=0.0758, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2528/2855 [1:44:21<10:41,  1.96s/it, loss=0.0758, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2528/2855 [1:44:23<10:41,  1.96s/it, loss=0.1890, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2529/2855 [1:44:23<10:54,  2.01s/it, loss=0.1890, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2529/2855 [1:44:25<10:54,  2.01s/it, loss=0.0878, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2530/2855 [1:44:25<10:34,  1.95s/it, loss=0.0878, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2530/2855 [1:44:27<10:34,  1.95s/it, loss=0.1103, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2531/2855 [1:44:27<10:32,  1.95s/it, loss=0.1103, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2531/2855 [1:44:29<10:32,  1.95s/it, loss=0.2106, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2532/2855 [1:44:29<10:39,  1.98s/it, loss=0.2106, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2532/2855 [1:44:31<10:39,  1.98s/it, loss=0.3013, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2533/2855 [1:44:31<10:26,  1.95s/it, loss=0.3013, lr=2.91e-05]
Epoch 1/8:  89%|████████▊ | 2533/2855 [1:44:32<10:26,  1.95s/it, loss=0.2304, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2534/2855 [1:44:32<10:17,  1.92s/it, loss=0.2304, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2534/2855 [1:44:34<10:17,  1.92s/it, loss=0.0954, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2535/2855 [1:44:34<09:41,  1.82s/it, loss=0.0954, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2535/2855 [1:44:36<09:41,  1.82s/it, loss=0.2278, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2536/2855 [1:44:36<09:26,  1.78s/it, loss=0.2278, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2536/2855 [1:44:38<09:26,  1.78s/it, loss=0.4188, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2537/2855 [1:44:38<09:59,  1.89s/it, loss=0.4188, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2537/2855 [1:44:40<09:59,  1.89s/it, loss=0.1173, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2538/2855 [1:44:40<09:56,  1.88s/it, loss=0.1173, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2538/2855 [1:44:42<09:56,  1.88s/it, loss=0.2563, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2539/2855 [1:44:42<10:34,  2.01s/it, loss=0.2563, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2539/2855 [1:44:44<10:34,  2.01s/it, loss=0.0888, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2540/2855 [1:44:44<10:25,  1.99s/it, loss=0.0888, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2540/2855 [1:44:46<10:25,  1.99s/it, loss=0.1569, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2541/2855 [1:44:46<10:01,  1.92s/it, loss=0.1569, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2541/2855 [1:44:48<10:01,  1.92s/it, loss=0.1380, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2542/2855 [1:44:48<10:35,  2.03s/it, loss=0.1380, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2542/2855 [1:44:50<10:35,  2.03s/it, loss=0.2723, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2543/2855 [1:44:50<10:38,  2.05s/it, loss=0.2723, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2543/2855 [1:44:52<10:38,  2.05s/it, loss=0.1620, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2544/2855 [1:44:52<10:36,  2.05s/it, loss=0.1620, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2544/2855 [1:44:54<10:36,  2.05s/it, loss=0.1614, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2545/2855 [1:44:54<10:29,  2.03s/it, loss=0.1614, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2545/2855 [1:44:56<10:29,  2.03s/it, loss=0.2409, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2546/2855 [1:44:56<10:16,  2.00s/it, loss=0.2409, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2546/2855 [1:44:58<10:16,  2.00s/it, loss=0.2269, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2547/2855 [1:44:58<10:23,  2.02s/it, loss=0.2269, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2547/2855 [1:45:00<10:23,  2.02s/it, loss=0.1067, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2548/2855 [1:45:00<10:14,  2.00s/it, loss=0.1067, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2548/2855 [1:45:02<10:14,  2.00s/it, loss=0.2428, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2549/2855 [1:45:02<10:21,  2.03s/it, loss=0.2428, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2549/2855 [1:45:04<10:21,  2.03s/it, loss=0.1375, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2550/2855 [1:45:04<10:13,  2.01s/it, loss=0.1375, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2550/2855 [1:45:06<10:13,  2.01s/it, loss=0.2112, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2551/2855 [1:45:06<09:56,  1.96s/it, loss=0.2112, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2551/2855 [1:45:08<09:56,  1.96s/it, loss=0.2616, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2552/2855 [1:45:08<10:15,  2.03s/it, loss=0.2616, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2552/2855 [1:45:10<10:15,  2.03s/it, loss=0.2184, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2553/2855 [1:45:10<10:09,  2.02s/it, loss=0.2184, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2553/2855 [1:45:12<10:09,  2.02s/it, loss=0.2038, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2554/2855 [1:45:12<10:05,  2.01s/it, loss=0.2038, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2554/2855 [1:45:14<10:05,  2.01s/it, loss=0.1394, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2555/2855 [1:45:14<09:39,  1.93s/it, loss=0.1394, lr=2.91e-05]
Epoch 1/8:  89%|████████▉ | 2555/2855 [1:45:15<09:39,  1.93s/it, loss=0.1891, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2556/2855 [1:45:15<08:58,  1.80s/it, loss=0.1891, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2556/2855 [1:45:18<08:58,  1.80s/it, loss=0.2074, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2557/2855 [1:45:18<09:57,  2.01s/it, loss=0.2074, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2557/2855 [1:45:20<09:57,  2.01s/it, loss=0.0786, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2558/2855 [1:45:20<09:53,  2.00s/it, loss=0.0786, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2558/2855 [1:45:22<09:53,  2.00s/it, loss=0.0964, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2559/2855 [1:45:22<09:44,  1.98s/it, loss=0.0964, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2559/2855 [1:45:24<09:44,  1.98s/it, loss=0.2014, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2560/2855 [1:45:24<10:03,  2.04s/it, loss=0.2014, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2560/2855 [1:45:26<10:03,  2.04s/it, loss=0.1333, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2561/2855 [1:45:26<09:43,  1.99s/it, loss=0.1333, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2561/2855 [1:45:28<09:43,  1.99s/it, loss=0.2257, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2562/2855 [1:45:28<09:42,  1.99s/it, loss=0.2257, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2562/2855 [1:45:30<09:42,  1.99s/it, loss=0.2817, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2563/2855 [1:45:30<09:47,  2.01s/it, loss=0.2817, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2563/2855 [1:45:32<09:47,  2.01s/it, loss=0.2842, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2564/2855 [1:45:32<09:39,  1.99s/it, loss=0.2842, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2564/2855 [1:45:34<09:39,  1.99s/it, loss=0.1615, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2565/2855 [1:45:34<09:39,  2.00s/it, loss=0.1615, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2565/2855 [1:45:36<09:39,  2.00s/it, loss=0.2950, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2566/2855 [1:45:36<09:17,  1.93s/it, loss=0.2950, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2566/2855 [1:45:38<09:17,  1.93s/it, loss=0.2222, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2567/2855 [1:45:38<09:18,  1.94s/it, loss=0.2222, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2567/2855 [1:45:40<09:18,  1.94s/it, loss=0.1169, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2568/2855 [1:45:40<09:27,  1.98s/it, loss=0.1169, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2568/2855 [1:45:41<09:27,  1.98s/it, loss=0.2832, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2569/2855 [1:45:41<09:18,  1.95s/it, loss=0.2832, lr=2.91e-05]
Epoch 1/8:  90%|████████▉ | 2569/2855 [1:45:43<09:18,  1.95s/it, loss=0.1724, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2570/2855 [1:45:43<09:13,  1.94s/it, loss=0.1724, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2570/2855 [1:45:45<09:13,  1.94s/it, loss=0.1798, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2571/2855 [1:45:45<08:36,  1.82s/it, loss=0.1798, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2571/2855 [1:45:47<08:36,  1.82s/it, loss=0.2352, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2572/2855 [1:45:47<08:22,  1.78s/it, loss=0.2352, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2572/2855 [1:45:49<08:22,  1.78s/it, loss=0.2452, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2573/2855 [1:45:49<08:52,  1.89s/it, loss=0.2452, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2573/2855 [1:45:51<08:52,  1.89s/it, loss=0.0699, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2574/2855 [1:45:51<09:04,  1.94s/it, loss=0.0699, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2574/2855 [1:45:53<09:04,  1.94s/it, loss=0.1925, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2575/2855 [1:45:53<09:11,  1.97s/it, loss=0.1925, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2575/2855 [1:45:55<09:11,  1.97s/it, loss=0.2892, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2576/2855 [1:45:55<09:14,  1.99s/it, loss=0.2892, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2576/2855 [1:45:57<09:14,  1.99s/it, loss=0.2063, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2577/2855 [1:45:57<08:58,  1.94s/it, loss=0.2063, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2577/2855 [1:45:59<08:58,  1.94s/it, loss=0.1765, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2578/2855 [1:45:59<09:01,  1.96s/it, loss=0.1765, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2578/2855 [1:46:01<09:01,  1.96s/it, loss=0.2375, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2579/2855 [1:46:01<08:56,  1.94s/it, loss=0.2375, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2579/2855 [1:46:02<08:56,  1.94s/it, loss=0.1996, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2580/2855 [1:46:02<08:48,  1.92s/it, loss=0.1996, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2580/2855 [1:46:04<08:48,  1.92s/it, loss=0.1784, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2581/2855 [1:46:04<08:45,  1.92s/it, loss=0.1784, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2581/2855 [1:46:06<08:45,  1.92s/it, loss=0.0795, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2582/2855 [1:46:06<08:29,  1.87s/it, loss=0.0795, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2582/2855 [1:46:08<08:29,  1.87s/it, loss=0.2409, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2583/2855 [1:46:08<08:36,  1.90s/it, loss=0.2409, lr=2.91e-05]
Epoch 1/8:  90%|█████████ | 2583/2855 [1:46:10<08:36,  1.90s/it, loss=0.1393, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2584/2855 [1:46:10<08:42,  1.93s/it, loss=0.1393, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2584/2855 [1:46:12<08:42,  1.93s/it, loss=0.1617, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2585/2855 [1:46:12<08:33,  1.90s/it, loss=0.1617, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2585/2855 [1:46:14<08:33,  1.90s/it, loss=0.2817, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2586/2855 [1:46:14<08:20,  1.86s/it, loss=0.2817, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2586/2855 [1:46:15<08:20,  1.86s/it, loss=0.1818, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2587/2855 [1:46:15<07:48,  1.75s/it, loss=0.1818, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2587/2855 [1:46:17<07:48,  1.75s/it, loss=0.0807, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2588/2855 [1:46:17<07:32,  1.69s/it, loss=0.0807, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2588/2855 [1:46:19<07:32,  1.69s/it, loss=0.1443, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2589/2855 [1:46:19<07:53,  1.78s/it, loss=0.1443, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2589/2855 [1:46:20<07:53,  1.78s/it, loss=0.0959, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2590/2855 [1:46:20<07:35,  1.72s/it, loss=0.0959, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2590/2855 [1:46:22<07:35,  1.72s/it, loss=0.1983, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2591/2855 [1:46:22<07:26,  1.69s/it, loss=0.1983, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2591/2855 [1:46:24<07:26,  1.69s/it, loss=0.1555, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2592/2855 [1:46:24<07:43,  1.76s/it, loss=0.1555, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2592/2855 [1:46:26<07:43,  1.76s/it, loss=0.3397, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2593/2855 [1:46:26<07:43,  1.77s/it, loss=0.3397, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2593/2855 [1:46:28<07:43,  1.77s/it, loss=0.1088, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2594/2855 [1:46:28<08:05,  1.86s/it, loss=0.1088, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2594/2855 [1:46:30<08:05,  1.86s/it, loss=0.2023, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2595/2855 [1:46:30<08:18,  1.92s/it, loss=0.2023, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2595/2855 [1:46:32<08:18,  1.92s/it, loss=0.1366, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2596/2855 [1:46:32<08:06,  1.88s/it, loss=0.1366, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2596/2855 [1:46:33<08:06,  1.88s/it, loss=0.0572, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2597/2855 [1:46:33<08:03,  1.87s/it, loss=0.0572, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2597/2855 [1:46:35<08:03,  1.87s/it, loss=0.1726, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2598/2855 [1:46:35<07:55,  1.85s/it, loss=0.1726, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2598/2855 [1:46:37<07:55,  1.85s/it, loss=0.1802, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2599/2855 [1:46:37<07:42,  1.81s/it, loss=0.1802, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2599/2855 [1:46:39<07:42,  1.81s/it, loss=0.1087, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2600/2855 [1:46:39<08:01,  1.89s/it, loss=0.1087, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2600/2855 [1:46:41<08:01,  1.89s/it, loss=0.3671, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2601/2855 [1:46:41<07:55,  1.87s/it, loss=0.3671, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2601/2855 [1:46:43<07:55,  1.87s/it, loss=0.2107, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2602/2855 [1:46:43<07:49,  1.86s/it, loss=0.2107, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2602/2855 [1:46:45<07:49,  1.86s/it, loss=0.1467, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2603/2855 [1:46:45<07:50,  1.87s/it, loss=0.1467, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2603/2855 [1:46:46<07:50,  1.87s/it, loss=0.1300, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2604/2855 [1:46:46<07:40,  1.83s/it, loss=0.1300, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2604/2855 [1:46:48<07:40,  1.83s/it, loss=0.2464, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2605/2855 [1:46:48<07:43,  1.85s/it, loss=0.2464, lr=2.91e-05]
Epoch 1/8:  91%|█████████ | 2605/2855 [1:46:50<07:43,  1.85s/it, loss=0.2510, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2606/2855 [1:46:50<07:28,  1.80s/it, loss=0.2510, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2606/2855 [1:46:51<07:28,  1.80s/it, loss=0.2906, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2607/2855 [1:46:51<07:05,  1.71s/it, loss=0.2906, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2607/2855 [1:46:53<07:05,  1.71s/it, loss=0.1967, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2608/2855 [1:46:53<07:14,  1.76s/it, loss=0.1967, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2608/2855 [1:46:55<07:14,  1.76s/it, loss=0.3314, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2609/2855 [1:46:55<07:35,  1.85s/it, loss=0.3314, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2609/2855 [1:46:57<07:35,  1.85s/it, loss=0.2157, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2610/2855 [1:46:57<07:28,  1.83s/it, loss=0.2157, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2610/2855 [1:46:59<07:28,  1.83s/it, loss=0.2396, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2611/2855 [1:46:59<07:30,  1.85s/it, loss=0.2396, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2611/2855 [1:47:01<07:30,  1.85s/it, loss=0.2553, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2612/2855 [1:47:01<07:42,  1.91s/it, loss=0.2553, lr=2.91e-05]
Epoch 1/8:  91%|█████████▏| 2612/2855 [1:47:03<07:42,  1.91s/it, loss=0.2458, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2613/2855 [1:47:03<07:42,  1.91s/it, loss=0.2458, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2613/2855 [1:47:05<07:42,  1.91s/it, loss=0.0405, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2614/2855 [1:47:05<07:32,  1.88s/it, loss=0.0405, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2614/2855 [1:47:07<07:32,  1.88s/it, loss=0.1415, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2615/2855 [1:47:07<07:34,  1.89s/it, loss=0.1415, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2615/2855 [1:47:09<07:34,  1.89s/it, loss=0.0943, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2616/2855 [1:47:09<08:19,  2.09s/it, loss=0.0943, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2616/2855 [1:47:13<08:19,  2.09s/it, loss=0.1032, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2617/2855 [1:47:13<10:00,  2.53s/it, loss=0.1032, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2617/2855 [1:47:16<10:00,  2.53s/it, loss=0.5969, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2618/2855 [1:47:16<10:35,  2.68s/it, loss=0.5969, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2618/2855 [1:47:19<10:35,  2.68s/it, loss=0.1563, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2619/2855 [1:47:19<11:09,  2.84s/it, loss=0.1563, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2619/2855 [1:47:22<11:09,  2.84s/it, loss=0.0827, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2620/2855 [1:47:22<11:38,  2.97s/it, loss=0.0827, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2620/2855 [1:47:25<11:38,  2.97s/it, loss=0.2584, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2621/2855 [1:47:25<11:48,  3.03s/it, loss=0.2584, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2621/2855 [1:47:28<11:48,  3.03s/it, loss=0.1907, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2622/2855 [1:47:28<11:43,  3.02s/it, loss=0.1907, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2622/2855 [1:47:32<11:43,  3.02s/it, loss=0.2063, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2623/2855 [1:47:32<11:59,  3.10s/it, loss=0.2063, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2623/2855 [1:47:35<11:59,  3.10s/it, loss=0.1557, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2624/2855 [1:47:35<11:57,  3.10s/it, loss=0.1557, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2624/2855 [1:47:38<11:57,  3.10s/it, loss=0.2465, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2625/2855 [1:47:38<11:47,  3.08s/it, loss=0.2465, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2625/2855 [1:47:41<11:47,  3.08s/it, loss=0.0057, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2626/2855 [1:47:41<11:58,  3.14s/it, loss=0.0057, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2626/2855 [1:47:44<11:58,  3.14s/it, loss=0.2708, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2627/2855 [1:47:44<12:00,  3.16s/it, loss=0.2708, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2627/2855 [1:47:47<12:00,  3.16s/it, loss=0.2652, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2628/2855 [1:47:47<11:47,  3.12s/it, loss=0.2652, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2628/2855 [1:47:51<11:47,  3.12s/it, loss=0.2327, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2629/2855 [1:47:51<11:57,  3.17s/it, loss=0.2327, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2629/2855 [1:47:54<11:57,  3.17s/it, loss=0.0726, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2630/2855 [1:47:54<11:54,  3.18s/it, loss=0.0726, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2630/2855 [1:47:57<11:54,  3.18s/it, loss=0.2704, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2631/2855 [1:47:57<11:42,  3.14s/it, loss=0.2704, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2631/2855 [1:48:00<11:42,  3.14s/it, loss=0.1727, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2632/2855 [1:48:00<11:41,  3.15s/it, loss=0.1727, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2632/2855 [1:48:03<11:41,  3.15s/it, loss=0.0394, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2633/2855 [1:48:03<11:48,  3.19s/it, loss=0.0394, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2633/2855 [1:48:06<11:48,  3.19s/it, loss=0.0882, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2634/2855 [1:48:06<11:33,  3.14s/it, loss=0.0882, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2634/2855 [1:48:09<11:33,  3.14s/it, loss=0.1213, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2635/2855 [1:48:09<11:21,  3.10s/it, loss=0.1213, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2635/2855 [1:48:13<11:21,  3.10s/it, loss=0.0933, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2636/2855 [1:48:13<11:28,  3.14s/it, loss=0.0933, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2636/2855 [1:48:16<11:28,  3.14s/it, loss=0.3410, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2637/2855 [1:48:16<11:18,  3.11s/it, loss=0.3410, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2637/2855 [1:48:19<11:18,  3.11s/it, loss=0.1251, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2638/2855 [1:48:19<11:11,  3.09s/it, loss=0.1251, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2638/2855 [1:48:22<11:11,  3.09s/it, loss=0.0812, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2639/2855 [1:48:22<11:12,  3.11s/it, loss=0.0812, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2639/2855 [1:48:25<11:12,  3.11s/it, loss=0.1665, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2640/2855 [1:48:25<11:08,  3.11s/it, loss=0.1665, lr=2.91e-05]
Epoch 1/8:  92%|█████████▏| 2640/2855 [1:48:28<11:08,  3.11s/it, loss=0.0745, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2641/2855 [1:48:28<11:05,  3.11s/it, loss=0.0745, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2641/2855 [1:48:31<11:05,  3.11s/it, loss=0.1950, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2642/2855 [1:48:31<11:05,  3.13s/it, loss=0.1950, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2642/2855 [1:48:35<11:05,  3.13s/it, loss=0.1158, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2643/2855 [1:48:35<11:08,  3.15s/it, loss=0.1158, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2643/2855 [1:48:38<11:08,  3.15s/it, loss=0.1173, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2644/2855 [1:48:38<10:55,  3.11s/it, loss=0.1173, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2644/2855 [1:48:41<10:55,  3.11s/it, loss=0.0838, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2645/2855 [1:48:41<10:50,  3.10s/it, loss=0.0838, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2645/2855 [1:48:44<10:50,  3.10s/it, loss=0.2403, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2646/2855 [1:48:44<10:57,  3.15s/it, loss=0.2403, lr=2.91e-05]
Epoch 1/8:  93%|█████████▎| 2646/2855 [1:48:47<10:57,  3.15s/it, loss=0.3970, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2647/2855 [1:48:47<10:50,  3.13s/it, loss=0.3970, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2647/2855 [1:48:50<10:50,  3.13s/it, loss=0.1102, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2648/2855 [1:48:50<10:40,  3.09s/it, loss=0.1102, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2648/2855 [1:48:53<10:40,  3.09s/it, loss=0.2090, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2649/2855 [1:48:53<10:45,  3.13s/it, loss=0.2090, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2649/2855 [1:48:56<10:45,  3.13s/it, loss=0.2323, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2650/2855 [1:48:56<10:29,  3.07s/it, loss=0.2323, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2650/2855 [1:48:59<10:29,  3.07s/it, loss=0.1861, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2651/2855 [1:48:59<10:22,  3.05s/it, loss=0.1861, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2651/2855 [1:49:02<10:22,  3.05s/it, loss=0.1115, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2652/2855 [1:49:02<10:28,  3.10s/it, loss=0.1115, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2652/2855 [1:49:06<10:28,  3.10s/it, loss=0.1410, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2653/2855 [1:49:06<10:31,  3.13s/it, loss=0.1410, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2653/2855 [1:49:08<10:31,  3.13s/it, loss=0.2439, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2654/2855 [1:49:08<10:20,  3.09s/it, loss=0.2439, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2654/2855 [1:49:12<10:20,  3.09s/it, loss=0.3372, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2655/2855 [1:49:12<10:19,  3.10s/it, loss=0.3372, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2655/2855 [1:49:15<10:19,  3.10s/it, loss=0.0590, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2656/2855 [1:49:15<10:25,  3.14s/it, loss=0.0590, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2656/2855 [1:49:18<10:25,  3.14s/it, loss=0.1970, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2657/2855 [1:49:18<10:13,  3.10s/it, loss=0.1970, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2657/2855 [1:49:21<10:13,  3.10s/it, loss=0.3549, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2658/2855 [1:49:21<10:17,  3.13s/it, loss=0.3549, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2658/2855 [1:49:24<10:17,  3.13s/it, loss=0.1512, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2659/2855 [1:49:24<10:18,  3.16s/it, loss=0.1512, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2659/2855 [1:49:27<10:18,  3.16s/it, loss=0.1222, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2660/2855 [1:49:27<10:04,  3.10s/it, loss=0.1222, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2660/2855 [1:49:30<10:04,  3.10s/it, loss=0.4704, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2661/2855 [1:49:30<09:59,  3.09s/it, loss=0.4704, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2661/2855 [1:49:34<09:59,  3.09s/it, loss=0.2109, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2662/2855 [1:49:34<10:04,  3.13s/it, loss=0.2109, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2662/2855 [1:49:37<10:04,  3.13s/it, loss=0.1102, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2663/2855 [1:49:37<09:53,  3.09s/it, loss=0.1102, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2663/2855 [1:49:40<09:53,  3.09s/it, loss=0.1626, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2664/2855 [1:49:40<09:44,  3.06s/it, loss=0.1626, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2664/2855 [1:49:43<09:44,  3.06s/it, loss=0.1442, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2665/2855 [1:49:43<09:45,  3.08s/it, loss=0.1442, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2665/2855 [1:49:46<09:45,  3.08s/it, loss=0.4110, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2666/2855 [1:49:46<09:48,  3.11s/it, loss=0.4110, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2666/2855 [1:49:49<09:48,  3.11s/it, loss=0.1613, lr=2.90e-05]
Epoch 1/8:  93%|█��███████▎| 2667/2855 [1:49:49<09:36,  3.07s/it, loss=0.1613, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2667/2855 [1:49:52<09:36,  3.07s/it, loss=0.1471, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2668/2855 [1:49:52<09:39,  3.10s/it, loss=0.1471, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2668/2855 [1:49:55<09:39,  3.10s/it, loss=0.1591, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2669/2855 [1:49:55<09:43,  3.14s/it, loss=0.1591, lr=2.90e-05]
Epoch 1/8:  93%|█████████▎| 2669/2855 [1:49:59<09:43,  3.14s/it, loss=0.2051, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2670/2855 [1:49:59<10:01,  3.25s/it, loss=0.2051, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2670/2855 [1:50:04<10:01,  3.25s/it, loss=0.1770, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2671/2855 [1:50:04<11:51,  3.87s/it, loss=0.1770, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2671/2855 [1:50:07<11:51,  3.87s/it, loss=0.3489, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2672/2855 [1:50:07<11:03,  3.62s/it, loss=0.3489, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2672/2855 [1:50:10<11:03,  3.62s/it, loss=0.2530, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2673/2855 [1:50:10<10:24,  3.43s/it, loss=0.2530, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2673/2855 [1:50:13<10:24,  3.43s/it, loss=0.2749, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2674/2855 [1:50:13<09:57,  3.30s/it, loss=0.2749, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2674/2855 [1:50:16<09:57,  3.30s/it, loss=0.3430, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2675/2855 [1:50:16<09:45,  3.25s/it, loss=0.3430, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2675/2855 [1:50:19<09:45,  3.25s/it, loss=0.2370, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2676/2855 [1:50:19<09:24,  3.15s/it, loss=0.2370, lr=2.90e-05]
Epoch 1/8:  94%|█████████▎| 2676/2855 [1:50:22<09:24,  3.15s/it, loss=0.2596, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2677/2855 [1:50:22<09:13,  3.11s/it, loss=0.2596, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2677/2855 [1:50:25<09:13,  3.11s/it, loss=0.1709, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2678/2855 [1:50:25<09:05,  3.08s/it, loss=0.1709, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2678/2855 [1:50:28<09:05,  3.08s/it, loss=0.2005, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2679/2855 [1:50:28<08:53,  3.03s/it, loss=0.2005, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2679/2855 [1:50:31<08:53,  3.03s/it, loss=0.0929, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2680/2855 [1:50:31<08:52,  3.04s/it, loss=0.0929, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2680/2855 [1:50:34<08:52,  3.04s/it, loss=0.0360, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2681/2855 [1:50:34<08:44,  3.02s/it, loss=0.0360, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2681/2855 [1:50:37<08:44,  3.02s/it, loss=0.2055, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2682/2855 [1:50:37<08:41,  3.01s/it, loss=0.2055, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2682/2855 [1:50:40<08:41,  3.01s/it, loss=0.0355, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2683/2855 [1:50:40<08:32,  2.98s/it, loss=0.0355, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2683/2855 [1:50:43<08:32,  2.98s/it, loss=0.1682, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2684/2855 [1:50:43<08:41,  3.05s/it, loss=0.1682, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2684/2855 [1:50:47<08:41,  3.05s/it, loss=0.2174, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2685/2855 [1:50:47<09:06,  3.21s/it, loss=0.2174, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2685/2855 [1:50:50<09:06,  3.21s/it, loss=0.3705, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2686/2855 [1:50:50<08:56,  3.18s/it, loss=0.3705, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2686/2855 [1:50:53<08:56,  3.18s/it, loss=0.0701, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2687/2855 [1:50:53<08:49,  3.15s/it, loss=0.0701, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2687/2855 [1:50:56<08:49,  3.15s/it, loss=0.2908, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2688/2855 [1:50:56<08:59,  3.23s/it, loss=0.2908, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2688/2855 [1:50:59<08:59,  3.23s/it, loss=0.1296, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2689/2855 [1:50:59<08:48,  3.18s/it, loss=0.1296, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2689/2855 [1:51:03<08:48,  3.18s/it, loss=0.1604, lr=2.90e-05]
Epoch 1/8:  94%|██████���██▍| 2690/2855 [1:51:03<08:39,  3.15s/it, loss=0.1604, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2690/2855 [1:51:06<08:39,  3.15s/it, loss=0.3758, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2691/2855 [1:51:06<08:38,  3.16s/it, loss=0.3758, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2691/2855 [1:51:09<08:38,  3.16s/it, loss=0.3127, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2692/2855 [1:51:09<08:34,  3.16s/it, loss=0.3127, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2692/2855 [1:51:12<08:34,  3.16s/it, loss=0.4181, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2693/2855 [1:51:12<08:26,  3.13s/it, loss=0.4181, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2693/2855 [1:51:15<08:26,  3.13s/it, loss=0.5552, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2694/2855 [1:51:15<08:44,  3.26s/it, loss=0.5552, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2694/2855 [1:51:19<08:44,  3.26s/it, loss=0.2445, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2695/2855 [1:51:19<08:39,  3.25s/it, loss=0.2445, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2695/2855 [1:51:22<08:39,  3.25s/it, loss=0.0963, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2696/2855 [1:51:22<08:24,  3.17s/it, loss=0.0963, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2696/2855 [1:51:25<08:24,  3.17s/it, loss=0.3813, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2697/2855 [1:51:25<08:20,  3.16s/it, loss=0.3813, lr=2.90e-05]
Epoch 1/8:  94%|█████████▍| 2697/2855 [1:51:28<08:20,  3.16s/it, loss=0.0962, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2698/2855 [1:51:28<08:20,  3.19s/it, loss=0.0962, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2698/2855 [1:51:31<08:20,  3.19s/it, loss=0.2457, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2699/2855 [1:51:31<08:10,  3.14s/it, loss=0.2457, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2699/2855 [1:51:34<08:10,  3.14s/it, loss=0.2562, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2700/2855 [1:51:34<08:06,  3.14s/it, loss=0.2562, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2700/2855 [1:51:38<08:06,  3.14s/it, loss=0.3503, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2701/2855 [1:51:38<08:10,  3.18s/it, loss=0.3503, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2701/2855 [1:51:41<08:10,  3.18s/it, loss=0.0753, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2702/2855 [1:51:41<07:58,  3.12s/it, loss=0.0753, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2702/2855 [1:51:44<07:58,  3.12s/it, loss=0.2097, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2703/2855 [1:51:44<07:54,  3.12s/it, loss=0.2097, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2703/2855 [1:51:47<07:54,  3.12s/it, loss=0.1488, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2704/2855 [1:51:47<07:59,  3.17s/it, loss=0.1488, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2704/2855 [1:51:50<07:59,  3.17s/it, loss=0.2445, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2705/2855 [1:51:50<07:50,  3.14s/it, loss=0.2445, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2705/2855 [1:51:53<07:50,  3.14s/it, loss=0.1653, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2706/2855 [1:51:53<07:50,  3.15s/it, loss=0.1653, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2706/2855 [1:51:56<07:50,  3.15s/it, loss=0.0107, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2707/2855 [1:51:56<07:47,  3.16s/it, loss=0.0107, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2707/2855 [1:52:00<07:47,  3.16s/it, loss=0.1791, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2708/2855 [1:52:00<07:47,  3.18s/it, loss=0.1791, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2708/2855 [1:52:03<07:47,  3.18s/it, loss=0.1776, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2709/2855 [1:52:03<07:40,  3.15s/it, loss=0.1776, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2709/2855 [1:52:06<07:40,  3.15s/it, loss=0.2163, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2710/2855 [1:52:06<07:27,  3.09s/it, loss=0.2163, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2710/2855 [1:52:09<07:27,  3.09s/it, loss=0.0739, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2711/2855 [1:52:09<07:22,  3.07s/it, loss=0.0739, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2711/2855 [1:52:12<07:22,  3.07s/it, loss=0.2295, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2712/2855 [1:52:12<07:16,  3.05s/it, loss=0.2295, lr=2.90e-05]
Epoch 1/8:  95%|█████████▍| 2712/2855 [1:52:15<07:16,  3.05s/it, loss=0.3197, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2713/2855 [1:52:15<07:18,  3.09s/it, loss=0.3197, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2713/2855 [1:52:18<07:18,  3.09s/it, loss=0.2491, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2714/2855 [1:52:18<07:10,  3.05s/it, loss=0.2491, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2714/2855 [1:52:21<07:10,  3.05s/it, loss=0.1587, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2715/2855 [1:52:21<07:04,  3.04s/it, loss=0.1587, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2715/2855 [1:52:24<07:04,  3.04s/it, loss=0.3868, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2716/2855 [1:52:24<07:01,  3.03s/it, loss=0.3868, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2716/2855 [1:52:27<07:01,  3.03s/it, loss=0.0480, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2717/2855 [1:52:27<06:58,  3.04s/it, loss=0.0480, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2717/2855 [1:52:28<06:58,  3.04s/it, loss=0.1078, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2718/2855 [1:52:28<05:41,  2.50s/it, loss=0.1078, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2718/2855 [1:52:30<05:41,  2.50s/it, loss=0.2255, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2719/2855 [1:52:30<05:00,  2.21s/it, loss=0.2255, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2719/2855 [1:52:31<05:00,  2.21s/it, loss=0.0997, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2720/2855 [1:52:31<04:32,  2.02s/it, loss=0.0997, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2720/2855 [1:52:33<04:32,  2.02s/it, loss=0.1314, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2721/2855 [1:52:33<04:05,  1.84s/it, loss=0.1314, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2721/2855 [1:52:34<04:05,  1.84s/it, loss=0.1930, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2722/2855 [1:52:34<03:47,  1.71s/it, loss=0.1930, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2722/2855 [1:52:36<03:47,  1.71s/it, loss=0.2773, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2723/2855 [1:52:36<03:37,  1.65s/it, loss=0.2773, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2723/2855 [1:52:37<03:37,  1.65s/it, loss=0.1964, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2724/2855 [1:52:37<03:26,  1.58s/it, loss=0.1964, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2724/2855 [1:52:38<03:26,  1.58s/it, loss=0.1277, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2725/2855 [1:52:38<03:14,  1.50s/it, loss=0.1277, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2725/2855 [1:52:40<03:14,  1.50s/it, loss=0.2074, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2726/2855 [1:52:40<03:06,  1.44s/it, loss=0.2074, lr=2.90e-05]
Epoch 1/8:  95%|█████████▌| 2726/2855 [1:52:41<03:06,  1.44s/it, loss=0.1561, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2727/2855 [1:52:41<02:58,  1.40s/it, loss=0.1561, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2727/2855 [1:52:42<02:58,  1.40s/it, loss=0.2258, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2728/2855 [1:52:42<02:51,  1.35s/it, loss=0.2258, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2728/2855 [1:52:43<02:51,  1.35s/it, loss=0.3944, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2729/2855 [1:52:43<02:47,  1.33s/it, loss=0.3944, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2729/2855 [1:52:45<02:47,  1.33s/it, loss=0.3170, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2730/2855 [1:52:45<02:50,  1.36s/it, loss=0.3170, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2730/2855 [1:52:46<02:50,  1.36s/it, loss=0.1096, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2731/2855 [1:52:46<02:56,  1.42s/it, loss=0.1096, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2731/2855 [1:52:48<02:56,  1.42s/it, loss=0.1781, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2732/2855 [1:52:48<02:58,  1.45s/it, loss=0.1781, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2732/2855 [1:52:50<02:58,  1.45s/it, loss=0.1486, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2733/2855 [1:52:50<03:05,  1.52s/it, loss=0.1486, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2733/2855 [1:52:51<03:05,  1.52s/it, loss=0.1935, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2734/2855 [1:52:51<03:01,  1.50s/it, loss=0.1935, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2734/2855 [1:52:52<03:01,  1.50s/it, loss=0.0400, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2735/2855 [1:52:52<02:56,  1.47s/it, loss=0.0400, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2735/2855 [1:52:54<02:56,  1.47s/it, loss=0.1498, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2736/2855 [1:52:54<02:56,  1.49s/it, loss=0.1498, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2736/2855 [1:52:56<02:56,  1.49s/it, loss=0.1821, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2737/2855 [1:52:56<02:59,  1.52s/it, loss=0.1821, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2737/2855 [1:52:57<02:59,  1.52s/it, loss=0.1500, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2738/2855 [1:52:57<02:54,  1.50s/it, loss=0.1500, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2738/2855 [1:52:58<02:54,  1.50s/it, loss=0.3080, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2739/2855 [1:52:58<02:49,  1.46s/it, loss=0.3080, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2739/2855 [1:53:00<02:49,  1.46s/it, loss=0.0560, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2740/2855 [1:53:00<02:48,  1.47s/it, loss=0.0560, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2740/2855 [1:53:01<02:48,  1.47s/it, loss=0.0673, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2741/2855 [1:53:01<02:48,  1.48s/it, loss=0.0673, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2741/2855 [1:53:03<02:48,  1.48s/it, loss=0.2313, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2742/2855 [1:53:03<02:46,  1.48s/it, loss=0.2313, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2742/2855 [1:53:05<02:46,  1.48s/it, loss=0.2578, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2743/2855 [1:53:05<02:53,  1.55s/it, loss=0.2578, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2743/2855 [1:53:06<02:53,  1.55s/it, loss=0.2248, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2744/2855 [1:53:06<02:56,  1.59s/it, loss=0.2248, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2744/2855 [1:53:08<02:56,  1.59s/it, loss=0.1188, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2745/2855 [1:53:08<02:52,  1.56s/it, loss=0.1188, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2745/2855 [1:53:09<02:52,  1.56s/it, loss=0.2152, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2746/2855 [1:53:09<02:45,  1.51s/it, loss=0.2152, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2746/2855 [1:53:11<02:45,  1.51s/it, loss=0.0584, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2747/2855 [1:53:11<02:42,  1.50s/it, loss=0.0584, lr=2.90e-05]
Epoch 1/8:  96%|█████████▌| 2747/2855 [1:53:12<02:42,  1.50s/it, loss=0.2087, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2748/2855 [1:53:12<02:36,  1.46s/it, loss=0.2087, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2748/2855 [1:53:13<02:36,  1.46s/it, loss=0.2041, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2749/2855 [1:53:13<02:29,  1.41s/it, loss=0.2041, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2749/2855 [1:53:15<02:29,  1.41s/it, loss=0.2066, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2750/2855 [1:53:15<02:24,  1.38s/it, loss=0.2066, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2750/2855 [1:53:16<02:24,  1.38s/it, loss=0.2747, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2751/2855 [1:53:16<02:24,  1.39s/it, loss=0.2747, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2751/2855 [1:53:17<02:24,  1.39s/it, loss=0.3673, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2752/2855 [1:53:17<02:22,  1.38s/it, loss=0.3673, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2752/2855 [1:53:19<02:22,  1.38s/it, loss=0.2633, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2753/2855 [1:53:19<02:19,  1.37s/it, loss=0.2633, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2753/2855 [1:53:20<02:19,  1.37s/it, loss=0.3051, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2754/2855 [1:53:20<02:20,  1.39s/it, loss=0.3051, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2754/2855 [1:53:22<02:20,  1.39s/it, loss=0.1583, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2755/2855 [1:53:22<02:24,  1.44s/it, loss=0.1583, lr=2.90e-05]
Epoch 1/8:  96%|█████████▋| 2755/2855 [1:53:23<02:24,  1.44s/it, loss=0.1304, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2756/2855 [1:53:23<02:25,  1.47s/it, loss=0.1304, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2756/2855 [1:53:25<02:25,  1.47s/it, loss=0.3056, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2757/2855 [1:53:25<02:28,  1.52s/it, loss=0.3056, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2757/2855 [1:53:27<02:28,  1.52s/it, loss=0.4297, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2758/2855 [1:53:27<02:31,  1.57s/it, loss=0.4297, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2758/2855 [1:53:28<02:31,  1.57s/it, loss=0.2076, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2759/2855 [1:53:28<02:31,  1.58s/it, loss=0.2076, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2759/2855 [1:53:30<02:31,  1.58s/it, loss=0.1665, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2760/2855 [1:53:30<02:33,  1.62s/it, loss=0.1665, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2760/2855 [1:53:32<02:33,  1.62s/it, loss=0.1730, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2761/2855 [1:53:32<02:33,  1.63s/it, loss=0.1730, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2761/2855 [1:53:33<02:33,  1.63s/it, loss=0.2992, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2762/2855 [1:53:33<02:28,  1.60s/it, loss=0.2992, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2762/2855 [1:53:35<02:28,  1.60s/it, loss=0.2272, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2763/2855 [1:53:35<02:27,  1.61s/it, loss=0.2272, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2763/2855 [1:53:37<02:27,  1.61s/it, loss=0.1808, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2764/2855 [1:53:37<02:31,  1.67s/it, loss=0.1808, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2764/2855 [1:53:38<02:31,  1.67s/it, loss=0.1291, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2765/2855 [1:53:38<02:28,  1.65s/it, loss=0.1291, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2765/2855 [1:53:40<02:28,  1.65s/it, loss=0.1088, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2766/2855 [1:53:40<02:25,  1.64s/it, loss=0.1088, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2766/2855 [1:53:41<02:25,  1.64s/it, loss=0.1450, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2767/2855 [1:53:41<02:27,  1.67s/it, loss=0.1450, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2767/2855 [1:53:43<02:27,  1.67s/it, loss=0.1783, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2768/2855 [1:53:43<02:23,  1.65s/it, loss=0.1783, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2768/2855 [1:53:45<02:23,  1.65s/it, loss=0.2257, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2769/2855 [1:53:45<02:16,  1.59s/it, loss=0.2257, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2769/2855 [1:53:46<02:16,  1.59s/it, loss=0.1748, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2770/2855 [1:53:46<02:13,  1.57s/it, loss=0.1748, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2770/2855 [1:53:48<02:13,  1.57s/it, loss=0.2269, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2771/2855 [1:53:48<02:10,  1.55s/it, loss=0.2269, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2771/2855 [1:53:49<02:10,  1.55s/it, loss=0.2754, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2772/2855 [1:53:49<02:05,  1.51s/it, loss=0.2754, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2772/2855 [1:53:51<02:05,  1.51s/it, loss=0.1380, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2773/2855 [1:53:51<02:08,  1.56s/it, loss=0.1380, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2773/2855 [1:53:52<02:08,  1.56s/it, loss=0.1436, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2774/2855 [1:53:52<02:06,  1.56s/it, loss=0.1436, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2774/2855 [1:53:54<02:06,  1.56s/it, loss=0.0082, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2775/2855 [1:53:54<02:02,  1.53s/it, loss=0.0082, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2775/2855 [1:53:55<02:02,  1.53s/it, loss=0.3215, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2776/2855 [1:53:55<02:02,  1.55s/it, loss=0.3215, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2776/2855 [1:53:57<02:02,  1.55s/it, loss=0.2566, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2777/2855 [1:53:57<02:02,  1.57s/it, loss=0.2566, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2777/2855 [1:53:58<02:02,  1.57s/it, loss=0.2501, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2778/2855 [1:53:58<01:59,  1.56s/it, loss=0.2501, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2778/2855 [1:54:00<01:59,  1.56s/it, loss=0.1505, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2779/2855 [1:54:00<01:57,  1.55s/it, loss=0.1505, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2779/2855 [1:54:02<01:57,  1.55s/it, loss=0.1318, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2780/2855 [1:54:02<02:00,  1.61s/it, loss=0.1318, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2780/2855 [1:54:03<02:00,  1.61s/it, loss=0.2384, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2781/2855 [1:54:03<01:56,  1.58s/it, loss=0.2384, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2781/2855 [1:54:05<01:56,  1.58s/it, loss=0.1791, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2782/2855 [1:54:05<01:53,  1.55s/it, loss=0.1791, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2782/2855 [1:54:06<01:53,  1.55s/it, loss=0.1168, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2783/2855 [1:54:06<01:54,  1.59s/it, loss=0.1168, lr=2.90e-05]
Epoch 1/8:  97%|█████████▋| 2783/2855 [1:54:08<01:54,  1.59s/it, loss=0.0626, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2784/2855 [1:54:08<02:03,  1.74s/it, loss=0.0626, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2784/2855 [1:54:11<02:03,  1.74s/it, loss=0.2547, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2785/2855 [1:54:11<02:09,  1.85s/it, loss=0.2547, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2785/2855 [1:54:12<02:09,  1.85s/it, loss=0.1392, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2786/2855 [1:54:12<02:06,  1.84s/it, loss=0.1392, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2786/2855 [1:54:14<02:06,  1.84s/it, loss=0.1330, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2787/2855 [1:54:14<02:01,  1.79s/it, loss=0.1330, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2787/2855 [1:54:16<02:01,  1.79s/it, loss=0.1411, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2788/2855 [1:54:16<02:01,  1.82s/it, loss=0.1411, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2788/2855 [1:54:18<02:01,  1.82s/it, loss=0.1419, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2789/2855 [1:54:18<02:03,  1.88s/it, loss=0.1419, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2789/2855 [1:54:20<02:03,  1.88s/it, loss=0.0743, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2790/2855 [1:54:20<02:02,  1.88s/it, loss=0.0743, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2790/2855 [1:54:22<02:02,  1.88s/it, loss=0.1934, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2791/2855 [1:54:22<02:03,  1.93s/it, loss=0.1934, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2791/2855 [1:54:24<02:03,  1.93s/it, loss=0.2099, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2792/2855 [1:54:24<02:00,  1.91s/it, loss=0.2099, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2792/2855 [1:54:26<02:00,  1.91s/it, loss=0.2355, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2793/2855 [1:54:26<01:57,  1.89s/it, loss=0.2355, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2793/2855 [1:54:28<01:57,  1.89s/it, loss=0.1188, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2794/2855 [1:54:28<01:58,  1.94s/it, loss=0.1188, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2794/2855 [1:54:30<01:58,  1.94s/it, loss=0.0582, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2795/2855 [1:54:30<01:55,  1.92s/it, loss=0.0582, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2795/2855 [1:54:31<01:55,  1.92s/it, loss=0.1751, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2796/2855 [1:54:31<01:54,  1.94s/it, loss=0.1751, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2796/2855 [1:54:33<01:54,  1.94s/it, loss=0.2088, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2797/2855 [1:54:33<01:47,  1.86s/it, loss=0.2088, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2797/2855 [1:54:35<01:47,  1.86s/it, loss=0.1727, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2798/2855 [1:54:35<01:40,  1.76s/it, loss=0.1727, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2798/2855 [1:54:37<01:40,  1.76s/it, loss=0.3963, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2799/2855 [1:54:37<01:42,  1.84s/it, loss=0.3963, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2799/2855 [1:54:39<01:42,  1.84s/it, loss=0.3394, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2800/2855 [1:54:39<01:42,  1.86s/it, loss=0.3394, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2800/2855 [1:54:40<01:42,  1.86s/it, loss=0.1707, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2801/2855 [1:54:40<01:39,  1.84s/it, loss=0.1707, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2801/2855 [1:54:43<01:39,  1.84s/it, loss=0.1900, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2802/2855 [1:54:43<01:41,  1.92s/it, loss=0.1900, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2802/2855 [1:54:44<01:41,  1.92s/it, loss=0.1926, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2803/2855 [1:54:44<01:40,  1.93s/it, loss=0.1926, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2803/2855 [1:54:47<01:40,  1.93s/it, loss=0.0630, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2804/2855 [1:54:47<01:53,  2.22s/it, loss=0.0630, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2804/2855 [1:54:49<01:53,  2.22s/it, loss=0.0837, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2805/2855 [1:54:49<01:48,  2.17s/it, loss=0.0837, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2805/2855 [1:54:52<01:48,  2.17s/it, loss=0.1924, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2806/2855 [1:54:52<01:46,  2.17s/it, loss=0.1924, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2806/2855 [1:54:54<01:46,  2.17s/it, loss=0.1885, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2807/2855 [1:54:54<01:44,  2.17s/it, loss=0.1885, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2807/2855 [1:54:56<01:44,  2.17s/it, loss=0.0010, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2808/2855 [1:54:56<01:40,  2.14s/it, loss=0.0010, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2808/2855 [1:54:58<01:40,  2.14s/it, loss=0.3514, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2809/2855 [1:54:58<01:39,  2.16s/it, loss=0.3514, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2809/2855 [1:55:00<01:39,  2.16s/it, loss=0.2424, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2810/2855 [1:55:00<01:36,  2.14s/it, loss=0.2424, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2810/2855 [1:55:02<01:36,  2.14s/it, loss=0.2152, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2811/2855 [1:55:02<01:36,  2.19s/it, loss=0.2152, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2811/2855 [1:55:04<01:36,  2.19s/it, loss=0.1226, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2812/2855 [1:55:04<01:32,  2.15s/it, loss=0.1226, lr=2.89e-05]
Epoch 1/8:  98%|█████████▊| 2812/2855 [1:55:07<01:32,  2.15s/it, loss=0.0766, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2813/2855 [1:55:07<01:31,  2.18s/it, loss=0.0766, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2813/2855 [1:55:09<01:31,  2.18s/it, loss=0.0621, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2814/2855 [1:55:09<01:30,  2.21s/it, loss=0.0621, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2814/2855 [1:55:11<01:30,  2.21s/it, loss=0.0891, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2815/2855 [1:55:11<01:28,  2.21s/it, loss=0.0891, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2815/2855 [1:55:13<01:28,  2.21s/it, loss=0.1877, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2816/2855 [1:55:13<01:24,  2.16s/it, loss=0.1877, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2816/2855 [1:55:15<01:24,  2.16s/it, loss=0.0998, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2817/2855 [1:55:15<01:18,  2.07s/it, loss=0.0998, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2817/2855 [1:55:17<01:18,  2.07s/it, loss=0.1416, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2818/2855 [1:55:17<01:17,  2.11s/it, loss=0.1416, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2818/2855 [1:55:19<01:17,  2.11s/it, loss=0.1154, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2819/2855 [1:55:19<01:14,  2.06s/it, loss=0.1154, lr=2.89e-05]
Epoch 1/8:  99%|█████████▊| 2819/2855 [1:55:22<01:14,  2.06s/it, loss=0.3634, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2820/2855 [1:55:22<01:14,  2.12s/it, loss=0.3634, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2820/2855 [1:55:26<01:14,  2.12s/it, loss=0.1488, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2821/2855 [1:55:26<01:35,  2.80s/it, loss=0.1488, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2821/2855 [1:55:51<01:35,  2.80s/it, loss=0.1990, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2822/2855 [1:55:51<05:11,  9.44s/it, loss=0.1990, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2822/2855 [1:55:52<05:11,  9.44s/it, loss=0.1654, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2823/2855 [1:55:52<03:46,  7.09s/it, loss=0.1654, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2823/2855 [1:55:55<03:46,  7.09s/it, loss=0.2693, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2824/2855 [1:55:55<02:53,  5.61s/it, loss=0.2693, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2824/2855 [1:55:57<02:53,  5.61s/it, loss=0.2860, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2825/2855 [1:55:57<02:18,  4.62s/it, loss=0.2860, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2825/2855 [1:55:59<02:18,  4.62s/it, loss=0.0931, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2826/2855 [1:55:59<01:49,  3.79s/it, loss=0.0931, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2826/2855 [1:56:01<01:49,  3.79s/it, loss=0.1824, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2827/2855 [1:56:01<01:34,  3.37s/it, loss=0.1824, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2827/2855 [1:56:04<01:34,  3.37s/it, loss=0.0755, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2828/2855 [1:56:04<01:23,  3.10s/it, loss=0.0755, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2828/2855 [1:56:06<01:23,  3.10s/it, loss=0.3931, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2829/2855 [1:56:06<01:12,  2.79s/it, loss=0.3931, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2829/2855 [1:56:08<01:12,  2.79s/it, loss=0.1143, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2830/2855 [1:56:08<01:05,  2.61s/it, loss=0.1143, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2830/2855 [1:56:10<01:05,  2.61s/it, loss=0.2441, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2831/2855 [1:56:10<01:00,  2.53s/it, loss=0.2441, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2831/2855 [1:56:13<01:00,  2.53s/it, loss=0.2509, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2832/2855 [1:56:13<00:57,  2.48s/it, loss=0.2509, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2832/2855 [1:56:15<00:57,  2.48s/it, loss=0.0968, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2833/2855 [1:56:15<00:51,  2.35s/it, loss=0.0968, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2833/2855 [1:56:17<00:51,  2.35s/it, loss=0.1984, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2834/2855 [1:56:17<00:47,  2.29s/it, loss=0.1984, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2834/2855 [1:56:19<00:47,  2.29s/it, loss=0.2974, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2835/2855 [1:56:19<00:45,  2.26s/it, loss=0.2974, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2835/2855 [1:56:21<00:45,  2.26s/it, loss=0.0023, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2836/2855 [1:56:21<00:43,  2.27s/it, loss=0.0023, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2836/2855 [1:56:24<00:43,  2.27s/it, loss=0.1338, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2837/2855 [1:56:24<00:41,  2.29s/it, loss=0.1338, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2837/2855 [1:56:26<00:41,  2.29s/it, loss=0.0796, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2838/2855 [1:56:26<00:37,  2.22s/it, loss=0.0796, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2838/2855 [1:56:28<00:37,  2.22s/it, loss=0.2866, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2839/2855 [1:56:28<00:35,  2.19s/it, loss=0.2866, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2839/2855 [1:56:30<00:35,  2.19s/it, loss=0.0750, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2840/2855 [1:56:30<00:31,  2.10s/it, loss=0.0750, lr=2.89e-05]
Epoch 1/8:  99%|█████████▉| 2840/2855 [1:56:32<00:31,  2.10s/it, loss=0.1154, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2841/2855 [1:56:32<00:30,  2.18s/it, loss=0.1154, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2841/2855 [1:56:34<00:30,  2.18s/it, loss=0.0814, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2842/2855 [1:56:34<00:27,  2.13s/it, loss=0.0814, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2842/2855 [1:56:36<00:27,  2.13s/it, loss=0.1036, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2843/2855 [1:56:36<00:25,  2.10s/it, loss=0.1036, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2843/2855 [1:56:39<00:25,  2.10s/it, loss=0.0717, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2844/2855 [1:56:39<00:24,  2.27s/it, loss=0.0717, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2844/2855 [1:56:41<00:24,  2.27s/it, loss=0.2592, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2845/2855 [1:56:41<00:22,  2.28s/it, loss=0.2592, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2845/2855 [1:56:43<00:22,  2.28s/it, loss=0.0988, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2846/2855 [1:56:43<00:20,  2.27s/it, loss=0.0988, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2846/2855 [1:56:45<00:20,  2.27s/it, loss=0.2230, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2847/2855 [1:56:45<00:17,  2.20s/it, loss=0.2230, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2847/2855 [1:56:47<00:17,  2.20s/it, loss=0.2329, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2848/2855 [1:56:47<00:15,  2.17s/it, loss=0.2329, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2848/2855 [1:56:50<00:15,  2.17s/it, loss=0.2142, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2849/2855 [1:56:50<00:12,  2.14s/it, loss=0.2142, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2849/2855 [1:56:52<00:12,  2.14s/it, loss=0.0244, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2850/2855 [1:56:52<00:11,  2.21s/it, loss=0.0244, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2850/2855 [1:56:54<00:11,  2.21s/it, loss=0.2738, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2851/2855 [1:56:54<00:08,  2.22s/it, loss=0.2738, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2851/2855 [1:56:56<00:08,  2.22s/it, loss=0.0411, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2852/2855 [1:56:56<00:06,  2.25s/it, loss=0.0411, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2852/2855 [1:56:59<00:06,  2.25s/it, loss=0.0861, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2853/2855 [1:56:59<00:04,  2.23s/it, loss=0.0861, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2853/2855 [1:57:01<00:04,  2.23s/it, loss=0.0514, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2854/2855 [1:57:01<00:02,  2.20s/it, loss=0.0514, lr=2.89e-05]
Epoch 1/8: 100%|█████████▉| 2854/2855 [1:57:03<00:02,  2.20s/it, loss=0.2096, lr=2.89e-05]
Epoch 1/8: 100%|██████████| 2855/2855 [1:57:03<00:00,  2.23s/it, loss=0.2096, lr=2.89e-05]
Epoch 1/8: 100%|██████████| 2855/2855 [1:57:03<00:00,  2.46s/it, loss=0.2096, lr=2.89e-05]
+
+✓ Epoch 1 | Avg Loss: 0.4221
+  💾 New best! Saving to checkpoints/oculus_reasoning_v2
+
Epoch 2/8:   0%|          | 0/2855 [00:00<?, ?it/s]
Epoch 2/8:   0%|          | 0/2855 [00:04<?, ?it/s, loss=0.1034, lr=2.89e-05]
Epoch 2/8:   0%|          | 1/2855 [00:04<3:42:04,  4.67s/it, loss=0.1034, lr=2.89e-05]
Epoch 2/8:   0%|          | 1/2855 [00:07<3:42:04,  4.67s/it, loss=0.0928, lr=2.89e-05]
Epoch 2/8:   0%|          | 2/2855 [00:07<2:59:29,  3.77s/it, loss=0.0928, lr=2.89e-05]
Epoch 2/8:   0%|          | 2/2855 [00:09<2:59:29,  3.77s/it, loss=0.1817, lr=2.89e-05]
Epoch 2/8:   0%|          | 3/2855 [00:09<2:21:06,  2.97s/it, loss=0.1817, lr=2.89e-05]
Epoch 2/8:   0%|          | 3/2855 [00:11<2:21:06,  2.97s/it, loss=0.0826, lr=2.89e-05]
Epoch 2/8:   0%|          | 4/2855 [00:11<2:05:12,  2.63s/it, loss=0.0826, lr=2.89e-05]
Epoch 2/8:   0%|          | 4/2855 [00:14<2:05:12,  2.63s/it, loss=0.0733, lr=2.89e-05]
Epoch 2/8:   0%|          | 5/2855 [00:14<1:58:46,  2.50s/it, loss=0.0733, lr=2.89e-05]
Epoch 2/8:   0%|          | 5/2855 [00:16<1:58:46,  2.50s/it, loss=0.0859, lr=2.89e-05]
Epoch 2/8:   0%|          | 6/2855 [00:16<1:52:57,  2.38s/it, loss=0.0859, lr=2.89e-05]
Epoch 2/8:   0%|          | 6/2855 [00:18<1:52:57,  2.38s/it, loss=0.1109, lr=2.89e-05]
Epoch 2/8:   0%|          | 7/2855 [00:18<1:50:14,  2.32s/it, loss=0.1109, lr=2.89e-05]
Epoch 2/8:   0%|          | 7/2855 [00:20<1:50:14,  2.32s/it, loss=0.1044, lr=2.89e-05]
Epoch 2/8:   0%|          | 8/2855 [00:20<1:47:04,  2.26s/it, loss=0.1044, lr=2.89e-05]
Epoch 2/8:   0%|          | 8/2855 [00:22<1:47:04,  2.26s/it, loss=0.1736, lr=2.89e-05]
Epoch 2/8:   0%|          | 9/2855 [00:22<1:44:39,  2.21s/it, loss=0.1736, lr=2.89e-05]
Epoch 2/8:   0%|          | 9/2855 [00:24<1:44:39,  2.21s/it, loss=0.1670, lr=2.89e-05]
Epoch 2/8:   0%|          | 10/2855 [00:24<1:44:47,  2.21s/it, loss=0.1670, lr=2.89e-05]
Epoch 2/8:   0%|          | 10/2855 [00:27<1:44:47,  2.21s/it, loss=0.0741, lr=2.89e-05]
Epoch 2/8:   0%|          | 11/2855 [00:27<1:44:53,  2.21s/it, loss=0.0741, lr=2.89e-05]
Epoch 2/8:   0%|          | 11/2855 [00:29<1:44:53,  2.21s/it, loss=0.2887, lr=2.89e-05]
Epoch 2/8:   0%|          | 12/2855 [00:29<1:44:33,  2.21s/it, loss=0.2887, lr=2.89e-05]
Epoch 2/8:   0%|          | 12/2855 [00:31<1:44:33,  2.21s/it, loss=0.0684, lr=2.89e-05]
Epoch 2/8:   0%|          | 13/2855 [00:31<1:43:24,  2.18s/it, loss=0.0684, lr=2.89e-05]
Epoch 2/8:   0%|          | 13/2855 [00:33<1:43:24,  2.18s/it, loss=0.0446, lr=2.89e-05]
Epoch 2/8:   0%|          | 14/2855 [00:33<1:43:07,  2.18s/it, loss=0.0446, lr=2.89e-05]
Epoch 2/8:   0%|          | 14/2855 [00:35<1:43:07,  2.18s/it, loss=0.1542, lr=2.89e-05]
Epoch 2/8:   1%|          | 15/2855 [00:35<1:44:11,  2.20s/it, loss=0.1542, lr=2.89e-05]
Epoch 2/8:   1%|          | 15/2855 [00:38<1:44:11,  2.20s/it, loss=0.2311, lr=2.89e-05]
Epoch 2/8:   1%|          | 16/2855 [00:38<1:45:09,  2.22s/it, loss=0.2311, lr=2.89e-05]
Epoch 2/8:   1%|          | 16/2855 [00:40<1:45:09,  2.22s/it, loss=0.1699, lr=2.89e-05]
Epoch 2/8:   1%|          | 17/2855 [00:40<1:43:33,  2.19s/it, loss=0.1699, lr=2.89e-05]
Epoch 2/8:   1%|          | 17/2855 [00:42<1:43:33,  2.19s/it, loss=0.2510, lr=2.89e-05]
Epoch 2/8:   1%|          | 18/2855 [00:42<1:42:02,  2.16s/it, loss=0.2510, lr=2.89e-05]
Epoch 2/8:   1%|          | 18/2855 [00:44<1:42:02,  2.16s/it, loss=0.3497, lr=2.89e-05]
Epoch 2/8:   1%|          | 19/2855 [00:44<1:42:02,  2.16s/it, loss=0.3497, lr=2.89e-05]
Epoch 2/8:   1%|          | 19/2855 [00:46<1:42:02,  2.16s/it, loss=0.2731, lr=2.89e-05]
Epoch 2/8:   1%|          | 20/2855 [00:46<1:43:38,  2.19s/it, loss=0.2731, lr=2.89e-05]
Epoch 2/8:   1%|          | 20/2855 [00:49<1:43:38,  2.19s/it, loss=0.2031, lr=2.89e-05]
Epoch 2/8:   1%|          | 21/2855 [00:49<1:44:11,  2.21s/it, loss=0.2031, lr=2.89e-05]
Epoch 2/8:   1%|          | 21/2855 [00:51<1:44:11,  2.21s/it, loss=0.0821, lr=2.89e-05]
Epoch 2/8:   1%|          | 22/2855 [00:51<1:42:35,  2.17s/it, loss=0.0821, lr=2.89e-05]
Epoch 2/8:   1%|          | 22/2855 [00:53<1:42:35,  2.17s/it, loss=0.2041, lr=2.89e-05]
Epoch 2/8:   1%|          | 23/2855 [00:53<1:40:49,  2.14s/it, loss=0.2041, lr=2.89e-05]
Epoch 2/8:   1%|          | 23/2855 [00:55<1:40:49,  2.14s/it, loss=0.1264, lr=2.89e-05]
Epoch 2/8:   1%|          | 24/2855 [00:55<1:41:04,  2.14s/it, loss=0.1264, lr=2.89e-05]
Epoch 2/8:   1%|          | 24/2855 [00:57<1:41:04,  2.14s/it, loss=0.0797, lr=2.89e-05]
Epoch 2/8:   1%|          | 25/2855 [00:57<1:41:37,  2.15s/it, loss=0.0797, lr=2.89e-05]
Epoch 2/8:   1%|          | 25/2855 [00:59<1:41:37,  2.15s/it, loss=0.0503, lr=2.89e-05]
Epoch 2/8:   1%|          | 26/2855 [00:59<1:41:27,  2.15s/it, loss=0.0503, lr=2.89e-05]
Epoch 2/8:   1%|          | 26/2855 [01:01<1:41:27,  2.15s/it, loss=0.0817, lr=2.89e-05]
Epoch 2/8:   1%|          | 27/2855 [01:01<1:39:33,  2.11s/it, loss=0.0817, lr=2.89e-05]
Epoch 2/8:   1%|          | 27/2855 [01:03<1:39:33,  2.11s/it, loss=0.2949, lr=2.89e-05]
Epoch 2/8:   1%|          | 28/2855 [01:03<1:39:56,  2.12s/it, loss=0.2949, lr=2.89e-05]
Epoch 2/8:   1%|          | 28/2855 [01:10<1:39:56,  2.12s/it, loss=0.0513, lr=2.89e-05]
Epoch 2/8:   1%|          | 29/2855 [01:10<2:38:38,  3.37s/it, loss=0.0513, lr=2.89e-05]
Epoch 2/8:   1%|          | 29/2855 [01:11<2:38:38,  3.37s/it, loss=0.0853, lr=2.89e-05]
Epoch 2/8:   1%|          | 30/2855 [01:11<2:10:38,  2.77s/it, loss=0.0853, lr=2.89e-05]
Epoch 2/8:   1%|          | 30/2855 [01:14<2:10:38,  2.77s/it, loss=0.1232, lr=2.89e-05]
Epoch 2/8:   1%|          | 31/2855 [01:14<2:13:20,  2.83s/it, loss=0.1232, lr=2.89e-05]
Epoch 2/8:   1%|          | 31/2855 [01:16<2:13:20,  2.83s/it, loss=0.1069, lr=2.89e-05]
Epoch 2/8:   1%|          | 32/2855 [01:16<2:01:35,  2.58s/it, loss=0.1069, lr=2.89e-05]
Epoch 2/8:   1%|          | 32/2855 [01:18<2:01:35,  2.58s/it, loss=0.2103, lr=2.89e-05]
Epoch 2/8:   1%|          | 33/2855 [01:18<1:56:05,  2.47s/it, loss=0.2103, lr=2.89e-05]
Epoch 2/8:   1%|          | 33/2855 [01:20<1:56:05,  2.47s/it, loss=0.1774, lr=2.89e-05]
Epoch 2/8:   1%|          | 34/2855 [01:20<1:51:17,  2.37s/it, loss=0.1774, lr=2.89e-05]
Epoch 2/8:   1%|          | 34/2855 [01:22<1:51:17,  2.37s/it, loss=0.1951, lr=2.89e-05]
Epoch 2/8:   1%|          | 35/2855 [01:22<1:47:49,  2.29s/it, loss=0.1951, lr=2.89e-05]
Epoch 2/8:   1%|          | 35/2855 [01:25<1:47:49,  2.29s/it, loss=0.1318, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 36/2855 [01:25<1:44:06,  2.22s/it, loss=0.1318, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 36/2855 [01:27<1:44:06,  2.22s/it, loss=0.1902, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 37/2855 [01:27<1:42:13,  2.18s/it, loss=0.1902, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 37/2855 [01:29<1:42:13,  2.18s/it, loss=0.0787, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 38/2855 [01:29<1:42:51,  2.19s/it, loss=0.0787, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 38/2855 [01:31<1:42:51,  2.19s/it, loss=0.2381, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 39/2855 [01:31<1:41:27,  2.16s/it, loss=0.2381, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 39/2855 [01:33<1:41:27,  2.16s/it, loss=0.1671, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 40/2855 [01:33<1:42:34,  2.19s/it, loss=0.1671, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 40/2855 [01:35<1:42:34,  2.19s/it, loss=0.1868, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 41/2855 [01:35<1:42:33,  2.19s/it, loss=0.1868, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 41/2855 [01:38<1:42:33,  2.19s/it, loss=0.1625, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 42/2855 [01:38<1:42:17,  2.18s/it, loss=0.1625, lr=2.89e-05]
Epoch 2/8:   1%|▏         | 42/2855 [01:40<1:42:17,  2.18s/it, loss=0.0820, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 43/2855 [01:40<1:41:32,  2.17s/it, loss=0.0820, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 43/2855 [01:42<1:41:32,  2.17s/it, loss=0.0476, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 44/2855 [01:42<1:39:42,  2.13s/it, loss=0.0476, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 44/2855 [01:44<1:39:42,  2.13s/it, loss=0.1422, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 45/2855 [01:44<1:39:08,  2.12s/it, loss=0.1422, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 45/2855 [01:46<1:39:08,  2.12s/it, loss=0.1275, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 46/2855 [01:46<1:39:08,  2.12s/it, loss=0.1275, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 46/2855 [01:48<1:39:08,  2.12s/it, loss=0.0862, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 47/2855 [01:48<1:40:06,  2.14s/it, loss=0.0862, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 47/2855 [01:50<1:40:06,  2.14s/it, loss=0.2102, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 48/2855 [01:50<1:40:04,  2.14s/it, loss=0.2102, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 48/2855 [01:52<1:40:04,  2.14s/it, loss=0.1274, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 49/2855 [01:52<1:37:45,  2.09s/it, loss=0.1274, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 49/2855 [01:54<1:37:45,  2.09s/it, loss=0.1912, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 50/2855 [01:54<1:37:21,  2.08s/it, loss=0.1912, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 50/2855 [01:56<1:37:21,  2.08s/it, loss=0.0072, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 51/2855 [01:56<1:38:19,  2.10s/it, loss=0.0072, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 51/2855 [01:59<1:38:19,  2.10s/it, loss=0.1430, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 52/2855 [01:59<1:40:55,  2.16s/it, loss=0.1430, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 52/2855 [02:01<1:40:55,  2.16s/it, loss=0.1709, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 53/2855 [02:01<1:40:12,  2.15s/it, loss=0.1709, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 53/2855 [02:03<1:40:12,  2.15s/it, loss=0.0711, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 54/2855 [02:03<1:37:36,  2.09s/it, loss=0.0711, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 54/2855 [02:05<1:37:36,  2.09s/it, loss=0.0430, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 55/2855 [02:05<1:39:42,  2.14s/it, loss=0.0430, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 55/2855 [02:07<1:39:42,  2.14s/it, loss=0.1677, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 56/2855 [02:07<1:40:16,  2.15s/it, loss=0.1677, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 56/2855 [02:09<1:40:16,  2.15s/it, loss=0.1644, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 57/2855 [02:09<1:39:18,  2.13s/it, loss=0.1644, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 57/2855 [02:11<1:39:18,  2.13s/it, loss=0.1951, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 58/2855 [02:11<1:38:48,  2.12s/it, loss=0.1951, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 58/2855 [02:13<1:38:48,  2.12s/it, loss=0.2560, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 59/2855 [02:13<1:36:31,  2.07s/it, loss=0.2560, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 59/2855 [02:15<1:36:31,  2.07s/it, loss=0.1552, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 60/2855 [02:15<1:37:33,  2.09s/it, loss=0.1552, lr=2.89e-05]
Epoch 2/8:   2%|▏         | 60/2855 [02:18<1:37:33,  2.09s/it, loss=0.1500, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 61/2855 [02:18<1:36:24,  2.07s/it, loss=0.1500, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 61/2855 [02:20<1:36:24,  2.07s/it, loss=0.1633, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 62/2855 [02:20<1:39:38,  2.14s/it, loss=0.1633, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 62/2855 [02:22<1:39:38,  2.14s/it, loss=0.1828, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 63/2855 [02:22<1:38:03,  2.11s/it, loss=0.1828, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 63/2855 [02:24<1:38:03,  2.11s/it, loss=0.2283, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 64/2855 [02:24<1:36:16,  2.07s/it, loss=0.2283, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 64/2855 [02:26<1:36:16,  2.07s/it, loss=0.0984, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 65/2855 [02:26<1:37:03,  2.09s/it, loss=0.0984, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 65/2855 [02:28<1:37:03,  2.09s/it, loss=0.1836, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 66/2855 [02:28<1:37:10,  2.09s/it, loss=0.1836, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 66/2855 [02:30<1:37:10,  2.09s/it, loss=0.2118, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 67/2855 [02:30<1:40:12,  2.16s/it, loss=0.2118, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 67/2855 [02:33<1:40:12,  2.16s/it, loss=0.1482, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 68/2855 [02:33<1:40:12,  2.16s/it, loss=0.1482, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 68/2855 [02:40<1:40:12,  2.16s/it, loss=0.1370, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 69/2855 [02:40<2:59:22,  3.86s/it, loss=0.1370, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 69/2855 [02:43<2:59:22,  3.86s/it, loss=0.1581, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 70/2855 [02:43<2:42:36,  3.50s/it, loss=0.1581, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 70/2855 [02:45<2:42:36,  3.50s/it, loss=0.1545, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 71/2855 [02:45<2:24:46,  3.12s/it, loss=0.1545, lr=2.88e-05]
Epoch 2/8:   2%|▏         | 71/2855 [02:47<2:24:46,  3.12s/it, loss=0.1461, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 72/2855 [02:47<2:09:43,  2.80s/it, loss=0.1461, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 72/2855 [02:49<2:09:43,  2.80s/it, loss=0.1933, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 73/2855 [02:49<2:00:58,  2.61s/it, loss=0.1933, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 73/2855 [02:52<2:00:58,  2.61s/it, loss=0.1276, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 74/2855 [02:52<1:53:48,  2.46s/it, loss=0.1276, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 74/2855 [02:54<1:53:48,  2.46s/it, loss=0.1501, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 75/2855 [02:54<1:46:48,  2.31s/it, loss=0.1501, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 75/2855 [02:56<1:46:48,  2.31s/it, loss=0.1286, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 76/2855 [02:56<1:44:39,  2.26s/it, loss=0.1286, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 76/2855 [02:58<1:44:39,  2.26s/it, loss=0.2202, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 77/2855 [02:58<1:42:12,  2.21s/it, loss=0.2202, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 77/2855 [03:00<1:42:12,  2.21s/it, loss=0.2798, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 78/2855 [03:00<1:41:29,  2.19s/it, loss=0.2798, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 78/2855 [03:02<1:41:29,  2.19s/it, loss=0.1067, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 79/2855 [03:02<1:41:47,  2.20s/it, loss=0.1067, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 79/2855 [03:05<1:41:47,  2.20s/it, loss=0.1089, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 80/2855 [03:05<1:45:07,  2.27s/it, loss=0.1089, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 80/2855 [03:07<1:45:07,  2.27s/it, loss=0.0353, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 81/2855 [03:07<1:49:42,  2.37s/it, loss=0.0353, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 81/2855 [03:09<1:49:42,  2.37s/it, loss=0.2083, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 82/2855 [03:09<1:48:09,  2.34s/it, loss=0.2083, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 82/2855 [03:12<1:48:09,  2.34s/it, loss=0.1596, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 83/2855 [03:12<1:45:55,  2.29s/it, loss=0.1596, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 83/2855 [03:14<1:45:55,  2.29s/it, loss=0.1469, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 84/2855 [03:14<1:42:25,  2.22s/it, loss=0.1469, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 84/2855 [03:16<1:42:25,  2.22s/it, loss=0.2431, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 85/2855 [03:16<1:38:42,  2.14s/it, loss=0.2431, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 85/2855 [03:18<1:38:42,  2.14s/it, loss=0.0407, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 86/2855 [03:18<1:38:57,  2.14s/it, loss=0.0407, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 86/2855 [03:20<1:38:57,  2.14s/it, loss=0.2106, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 87/2855 [03:20<1:38:39,  2.14s/it, loss=0.2106, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 87/2855 [03:22<1:38:39,  2.14s/it, loss=0.0629, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 88/2855 [03:22<1:37:00,  2.10s/it, loss=0.0629, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 88/2855 [03:24<1:37:00,  2.10s/it, loss=0.1480, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 89/2855 [03:24<1:35:05,  2.06s/it, loss=0.1480, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 89/2855 [03:26<1:35:05,  2.06s/it, loss=0.1140, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 90/2855 [03:26<1:34:27,  2.05s/it, loss=0.1140, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 90/2855 [03:28<1:34:27,  2.05s/it, loss=0.1605, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 91/2855 [03:28<1:33:01,  2.02s/it, loss=0.1605, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 91/2855 [03:30<1:33:01,  2.02s/it, loss=0.1731, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 92/2855 [03:30<1:34:38,  2.06s/it, loss=0.1731, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 92/2855 [03:32<1:34:38,  2.06s/it, loss=0.2328, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 93/2855 [03:32<1:32:41,  2.01s/it, loss=0.2328, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 93/2855 [03:34<1:32:41,  2.01s/it, loss=0.1598, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 94/2855 [03:34<1:32:50,  2.02s/it, loss=0.1598, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 94/2855 [03:36<1:32:50,  2.02s/it, loss=0.0355, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 95/2855 [03:36<1:33:41,  2.04s/it, loss=0.0355, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 95/2855 [03:38<1:33:41,  2.04s/it, loss=0.1732, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 96/2855 [03:38<1:31:59,  2.00s/it, loss=0.1732, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 96/2855 [03:40<1:31:59,  2.00s/it, loss=0.1858, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 97/2855 [03:40<1:33:18,  2.03s/it, loss=0.1858, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 97/2855 [03:42<1:33:18,  2.03s/it, loss=0.0960, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 98/2855 [03:42<1:30:24,  1.97s/it, loss=0.0960, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 98/2855 [03:44<1:30:24,  1.97s/it, loss=0.1389, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 99/2855 [03:44<1:28:10,  1.92s/it, loss=0.1389, lr=2.88e-05]
Epoch 2/8:   3%|▎         | 99/2855 [03:46<1:28:10,  1.92s/it, loss=0.1173, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 100/2855 [03:46<1:34:10,  2.05s/it, loss=0.1173, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 100/2855 [03:48<1:34:10,  2.05s/it, loss=0.3095, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 101/2855 [03:48<1:34:13,  2.05s/it, loss=0.3095, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 101/2855 [03:50<1:34:13,  2.05s/it, loss=0.1020, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 102/2855 [03:50<1:35:18,  2.08s/it, loss=0.1020, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 102/2855 [03:52<1:35:18,  2.08s/it, loss=0.1685, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 103/2855 [03:52<1:33:26,  2.04s/it, loss=0.1685, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 103/2855 [03:54<1:33:26,  2.04s/it, loss=0.0975, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 104/2855 [03:54<1:31:39,  2.00s/it, loss=0.0975, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 104/2855 [03:56<1:31:39,  2.00s/it, loss=0.1377, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 105/2855 [03:56<1:31:05,  1.99s/it, loss=0.1377, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 105/2855 [03:58<1:31:05,  1.99s/it, loss=0.1295, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 106/2855 [03:58<1:30:41,  1.98s/it, loss=0.1295, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 106/2855 [04:00<1:30:41,  1.98s/it, loss=0.0786, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 107/2855 [04:00<1:30:22,  1.97s/it, loss=0.0786, lr=2.88e-05]
Epoch 2/8:   4%|▎         | 107/2855 [04:06<1:30:22,  1.97s/it, loss=0.2832, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 108/2855 [04:06<2:25:26,  3.18s/it, loss=0.2832, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 108/2855 [04:07<2:25:26,  3.18s/it, loss=0.2840, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 109/2855 [04:07<1:59:23,  2.61s/it, loss=0.2840, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 109/2855 [04:10<1:59:23,  2.61s/it, loss=0.1106, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 110/2855 [04:10<1:55:29,  2.52s/it, loss=0.1106, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 110/2855 [04:12<1:55:29,  2.52s/it, loss=0.2139, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 111/2855 [04:12<1:49:56,  2.40s/it, loss=0.2139, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 111/2855 [04:13<1:49:56,  2.40s/it, loss=0.0944, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 112/2855 [04:13<1:40:39,  2.20s/it, loss=0.0944, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 112/2855 [04:15<1:40:39,  2.20s/it, loss=0.1116, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 113/2855 [04:15<1:35:23,  2.09s/it, loss=0.1116, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 113/2855 [04:17<1:35:23,  2.09s/it, loss=0.0601, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 114/2855 [04:17<1:31:05,  1.99s/it, loss=0.0601, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 114/2855 [04:19<1:31:05,  1.99s/it, loss=0.1380, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 115/2855 [04:19<1:36:08,  2.11s/it, loss=0.1380, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 115/2855 [04:21<1:36:08,  2.11s/it, loss=0.0641, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 116/2855 [04:21<1:35:38,  2.10s/it, loss=0.0641, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 116/2855 [04:23<1:35:38,  2.10s/it, loss=0.1823, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 117/2855 [04:23<1:31:20,  2.00s/it, loss=0.1823, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 117/2855 [04:25<1:31:20,  2.00s/it, loss=0.0540, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 118/2855 [04:25<1:29:03,  1.95s/it, loss=0.0540, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 118/2855 [04:27<1:29:03,  1.95s/it, loss=0.1551, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 119/2855 [04:27<1:29:57,  1.97s/it, loss=0.1551, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 119/2855 [04:29<1:29:57,  1.97s/it, loss=0.0362, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 120/2855 [04:29<1:29:27,  1.96s/it, loss=0.0362, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 120/2855 [04:31<1:29:27,  1.96s/it, loss=0.1388, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 121/2855 [04:31<1:29:17,  1.96s/it, loss=0.1388, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 121/2855 [04:33<1:29:17,  1.96s/it, loss=0.1745, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 122/2855 [04:33<1:28:04,  1.93s/it, loss=0.1745, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 122/2855 [04:35<1:28:04,  1.93s/it, loss=0.2216, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 123/2855 [04:35<1:27:29,  1.92s/it, loss=0.2216, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 123/2855 [04:37<1:27:29,  1.92s/it, loss=0.0562, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 124/2855 [04:37<1:26:24,  1.90s/it, loss=0.0562, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 124/2855 [04:38<1:26:24,  1.90s/it, loss=0.1698, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 125/2855 [04:38<1:25:18,  1.87s/it, loss=0.1698, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 125/2855 [04:40<1:25:18,  1.87s/it, loss=0.0538, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 126/2855 [04:40<1:27:22,  1.92s/it, loss=0.0538, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 126/2855 [04:42<1:27:22,  1.92s/it, loss=0.1082, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 127/2855 [04:42<1:28:51,  1.95s/it, loss=0.1082, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 127/2855 [04:44<1:28:51,  1.95s/it, loss=0.1090, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 128/2855 [04:44<1:26:43,  1.91s/it, loss=0.1090, lr=2.88e-05]
Epoch 2/8:   4%|▍         | 128/2855 [04:46<1:26:43,  1.91s/it, loss=0.1843, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 129/2855 [04:46<1:25:38,  1.89s/it, loss=0.1843, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 129/2855 [04:48<1:25:38,  1.89s/it, loss=0.0271, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 130/2855 [04:48<1:26:35,  1.91s/it, loss=0.0271, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 130/2855 [04:50<1:26:35,  1.91s/it, loss=0.1617, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 131/2855 [04:50<1:27:56,  1.94s/it, loss=0.1617, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 131/2855 [04:52<1:27:56,  1.94s/it, loss=0.2118, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 132/2855 [04:52<1:27:46,  1.93s/it, loss=0.2118, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 132/2855 [04:54<1:27:46,  1.93s/it, loss=0.0905, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 133/2855 [04:54<1:23:41,  1.84s/it, loss=0.0905, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 133/2855 [04:55<1:23:41,  1.84s/it, loss=0.1143, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 134/2855 [04:55<1:23:53,  1.85s/it, loss=0.1143, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 134/2855 [04:58<1:23:53,  1.85s/it, loss=0.0900, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 135/2855 [04:58<1:26:41,  1.91s/it, loss=0.0900, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 135/2855 [04:59<1:26:41,  1.91s/it, loss=0.0996, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 136/2855 [04:59<1:27:08,  1.92s/it, loss=0.0996, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 136/2855 [05:01<1:27:08,  1.92s/it, loss=0.0301, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 137/2855 [05:01<1:28:04,  1.94s/it, loss=0.0301, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 137/2855 [05:03<1:28:04,  1.94s/it, loss=0.1587, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 138/2855 [05:03<1:27:27,  1.93s/it, loss=0.1587, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 138/2855 [05:05<1:27:27,  1.93s/it, loss=0.3842, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 139/2855 [05:05<1:26:50,  1.92s/it, loss=0.3842, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 139/2855 [05:07<1:26:50,  1.92s/it, loss=0.1547, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 140/2855 [05:07<1:25:50,  1.90s/it, loss=0.1547, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 140/2855 [05:09<1:25:50,  1.90s/it, loss=0.2372, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 141/2855 [05:09<1:26:00,  1.90s/it, loss=0.2372, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 141/2855 [05:11<1:26:00,  1.90s/it, loss=0.1324, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 142/2855 [05:11<1:26:57,  1.92s/it, loss=0.1324, lr=2.88e-05]
Epoch 2/8:   5%|▍         | 142/2855 [05:13<1:26:57,  1.92s/it, loss=0.0165, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 143/2855 [05:13<1:25:21,  1.89s/it, loss=0.0165, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 143/2855 [05:15<1:25:21,  1.89s/it, loss=0.1180, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 144/2855 [05:15<1:24:59,  1.88s/it, loss=0.1180, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 144/2855 [05:17<1:24:59,  1.88s/it, loss=0.1913, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 145/2855 [05:17<1:24:34,  1.87s/it, loss=0.1913, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 145/2855 [05:18<1:24:34,  1.87s/it, loss=0.1100, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 146/2855 [05:18<1:24:28,  1.87s/it, loss=0.1100, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 146/2855 [05:20<1:24:28,  1.87s/it, loss=0.1164, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 147/2855 [05:20<1:24:40,  1.88s/it, loss=0.1164, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 147/2855 [05:26<1:24:40,  1.88s/it, loss=0.2259, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 148/2855 [05:26<2:15:09,  3.00s/it, loss=0.2259, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 148/2855 [05:27<2:15:09,  3.00s/it, loss=0.2043, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 149/2855 [05:27<1:51:53,  2.48s/it, loss=0.2043, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 149/2855 [05:29<1:51:53,  2.48s/it, loss=0.2761, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 150/2855 [05:29<1:40:03,  2.22s/it, loss=0.2761, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 150/2855 [05:31<1:40:03,  2.22s/it, loss=0.1112, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 151/2855 [05:31<1:39:48,  2.21s/it, loss=0.1112, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 151/2855 [05:33<1:39:48,  2.21s/it, loss=0.3063, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 152/2855 [05:33<1:42:33,  2.28s/it, loss=0.3063, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 152/2855 [05:41<1:42:33,  2.28s/it, loss=0.0813, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 153/2855 [05:41<2:49:04,  3.75s/it, loss=0.0813, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 153/2855 [05:48<2:49:04,  3.75s/it, loss=0.1304, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 154/2855 [05:48<3:31:36,  4.70s/it, loss=0.1304, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 154/2855 [05:50<3:31:36,  4.70s/it, loss=0.1133, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 155/2855 [05:50<3:04:16,  4.10s/it, loss=0.1133, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 155/2855 [05:52<3:04:16,  4.10s/it, loss=0.1446, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 156/2855 [05:52<2:36:08,  3.47s/it, loss=0.1446, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 156/2855 [05:54<2:36:08,  3.47s/it, loss=0.0866, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 157/2855 [05:54<2:14:29,  2.99s/it, loss=0.0866, lr=2.88e-05]
Epoch 2/8:   5%|▌         | 157/2855 [05:56<2:14:29,  2.99s/it, loss=0.2056, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 158/2855 [05:56<2:04:27,  2.77s/it, loss=0.2056, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 158/2855 [05:58<2:04:27,  2.77s/it, loss=0.1163, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 159/2855 [05:58<1:53:53,  2.53s/it, loss=0.1163, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 159/2855 [06:00<1:53:53,  2.53s/it, loss=0.1237, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 160/2855 [06:00<1:46:29,  2.37s/it, loss=0.1237, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 160/2855 [06:02<1:46:29,  2.37s/it, loss=0.1741, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 161/2855 [06:02<1:38:39,  2.20s/it, loss=0.1741, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 161/2855 [06:04<1:38:39,  2.20s/it, loss=0.2277, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 162/2855 [06:04<1:34:41,  2.11s/it, loss=0.2277, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 162/2855 [06:06<1:34:41,  2.11s/it, loss=0.1461, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 163/2855 [06:06<1:34:01,  2.10s/it, loss=0.1461, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 163/2855 [06:08<1:34:01,  2.10s/it, loss=0.0596, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 164/2855 [06:08<1:31:46,  2.05s/it, loss=0.0596, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 164/2855 [06:10<1:31:46,  2.05s/it, loss=0.2326, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 165/2855 [06:10<1:33:45,  2.09s/it, loss=0.2326, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 165/2855 [06:12<1:33:45,  2.09s/it, loss=0.0703, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 166/2855 [06:12<1:32:03,  2.05s/it, loss=0.0703, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 166/2855 [06:14<1:32:03,  2.05s/it, loss=0.1814, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 167/2855 [06:14<1:30:19,  2.02s/it, loss=0.1814, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 167/2855 [06:16<1:30:19,  2.02s/it, loss=0.2409, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 168/2855 [06:16<1:32:11,  2.06s/it, loss=0.2409, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 168/2855 [06:18<1:32:11,  2.06s/it, loss=0.1360, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 169/2855 [06:18<1:34:26,  2.11s/it, loss=0.1360, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 169/2855 [06:21<1:34:26,  2.11s/it, loss=0.1631, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 170/2855 [06:21<1:35:25,  2.13s/it, loss=0.1631, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 170/2855 [06:23<1:35:25,  2.13s/it, loss=0.1049, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 171/2855 [06:23<1:32:57,  2.08s/it, loss=0.1049, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 171/2855 [06:25<1:32:57,  2.08s/it, loss=0.2259, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 172/2855 [06:25<1:33:31,  2.09s/it, loss=0.2259, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 172/2855 [06:27<1:33:31,  2.09s/it, loss=0.1321, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 173/2855 [06:27<1:34:14,  2.11s/it, loss=0.1321, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 173/2855 [06:29<1:34:14,  2.11s/it, loss=0.1796, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 174/2855 [06:29<1:31:58,  2.06s/it, loss=0.1796, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 174/2855 [06:31<1:31:58,  2.06s/it, loss=0.2390, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 175/2855 [06:31<1:34:07,  2.11s/it, loss=0.2390, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 175/2855 [06:33<1:34:07,  2.11s/it, loss=0.1410, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 176/2855 [06:33<1:32:25,  2.07s/it, loss=0.1410, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 176/2855 [06:35<1:32:25,  2.07s/it, loss=0.1161, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 177/2855 [06:35<1:37:22,  2.18s/it, loss=0.1161, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 177/2855 [06:37<1:37:22,  2.18s/it, loss=0.1254, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 178/2855 [06:37<1:33:54,  2.10s/it, loss=0.1254, lr=2.88e-05]
Epoch 2/8:   6%|▌         | 178/2855 [06:39<1:33:54,  2.10s/it, loss=0.2104, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 179/2855 [06:39<1:33:38,  2.10s/it, loss=0.2104, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 179/2855 [06:42<1:33:38,  2.10s/it, loss=0.1932, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 180/2855 [06:42<1:32:59,  2.09s/it, loss=0.1932, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 180/2855 [06:43<1:32:59,  2.09s/it, loss=0.1567, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 181/2855 [06:43<1:29:47,  2.01s/it, loss=0.1567, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 181/2855 [06:45<1:29:47,  2.01s/it, loss=0.2049, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 182/2855 [06:45<1:30:48,  2.04s/it, loss=0.2049, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 182/2855 [06:47<1:30:48,  2.04s/it, loss=0.1646, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 183/2855 [06:47<1:28:49,  1.99s/it, loss=0.1646, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 183/2855 [06:49<1:28:49,  1.99s/it, loss=0.1458, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 184/2855 [06:49<1:27:46,  1.97s/it, loss=0.1458, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 184/2855 [06:51<1:27:46,  1.97s/it, loss=0.1508, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 185/2855 [06:51<1:29:01,  2.00s/it, loss=0.1508, lr=2.88e-05]
Epoch 2/8:   6%|▋         | 185/2855 [06:53<1:29:01,  2.00s/it, loss=0.1499, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 186/2855 [06:53<1:27:52,  1.98s/it, loss=0.1499, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 186/2855 [06:56<1:27:52,  1.98s/it, loss=0.2821, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 187/2855 [06:56<1:31:34,  2.06s/it, loss=0.2821, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 187/2855 [07:02<1:31:34,  2.06s/it, loss=0.2093, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 188/2855 [07:02<2:35:33,  3.50s/it, loss=0.2093, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 188/2855 [07:04<2:35:33,  3.50s/it, loss=0.1458, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 189/2855 [07:04<2:05:50,  2.83s/it, loss=0.1458, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 189/2855 [07:06<2:05:50,  2.83s/it, loss=0.1084, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 190/2855 [07:06<1:53:37,  2.56s/it, loss=0.1084, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 190/2855 [07:08<1:53:37,  2.56s/it, loss=0.2186, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 191/2855 [07:08<1:53:00,  2.55s/it, loss=0.2186, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 191/2855 [07:10<1:53:00,  2.55s/it, loss=0.2161, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 192/2855 [07:10<1:45:18,  2.37s/it, loss=0.2161, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 192/2855 [07:12<1:45:18,  2.37s/it, loss=0.0606, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 193/2855 [07:12<1:39:24,  2.24s/it, loss=0.0606, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 193/2855 [07:14<1:39:24,  2.24s/it, loss=0.1209, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 194/2855 [07:14<1:36:12,  2.17s/it, loss=0.1209, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 194/2855 [07:16<1:36:12,  2.17s/it, loss=0.0523, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 195/2855 [07:16<1:36:44,  2.18s/it, loss=0.0523, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 195/2855 [07:18<1:36:44,  2.18s/it, loss=0.1358, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 196/2855 [07:18<1:33:55,  2.12s/it, loss=0.1358, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 196/2855 [07:20<1:33:55,  2.12s/it, loss=0.0790, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 197/2855 [07:20<1:32:41,  2.09s/it, loss=0.0790, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 197/2855 [07:22<1:32:41,  2.09s/it, loss=0.2331, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 198/2855 [07:22<1:34:57,  2.14s/it, loss=0.2331, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 198/2855 [07:24<1:34:57,  2.14s/it, loss=0.1519, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 199/2855 [07:24<1:32:21,  2.09s/it, loss=0.1519, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 199/2855 [07:27<1:32:21,  2.09s/it, loss=0.1013, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 200/2855 [07:27<1:35:45,  2.16s/it, loss=0.1013, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 200/2855 [07:29<1:35:45,  2.16s/it, loss=0.0430, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 201/2855 [07:29<1:35:52,  2.17s/it, loss=0.0430, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 201/2855 [07:31<1:35:52,  2.17s/it, loss=0.1322, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 202/2855 [07:31<1:37:53,  2.21s/it, loss=0.1322, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 202/2855 [07:34<1:37:53,  2.21s/it, loss=0.3556, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 203/2855 [07:34<1:40:05,  2.26s/it, loss=0.3556, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 203/2855 [07:36<1:40:05,  2.26s/it, loss=0.1730, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 204/2855 [07:36<1:41:09,  2.29s/it, loss=0.1730, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 204/2855 [07:38<1:41:09,  2.29s/it, loss=0.2667, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 205/2855 [07:38<1:39:05,  2.24s/it, loss=0.2667, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 205/2855 [07:40<1:39:05,  2.24s/it, loss=0.1287, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 206/2855 [07:40<1:38:00,  2.22s/it, loss=0.1287, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 206/2855 [07:42<1:38:00,  2.22s/it, loss=0.1432, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 207/2855 [07:42<1:37:02,  2.20s/it, loss=0.1432, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 207/2855 [07:45<1:37:02,  2.20s/it, loss=0.1498, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 208/2855 [07:45<1:38:16,  2.23s/it, loss=0.1498, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 208/2855 [07:47<1:38:16,  2.23s/it, loss=0.1160, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 209/2855 [07:47<1:40:46,  2.29s/it, loss=0.1160, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 209/2855 [07:51<1:40:46,  2.29s/it, loss=0.0460, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 210/2855 [07:51<1:56:17,  2.64s/it, loss=0.0460, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 210/2855 [07:56<1:56:17,  2.64s/it, loss=0.1218, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 211/2855 [07:56<2:37:23,  3.57s/it, loss=0.1218, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 211/2855 [08:02<2:37:23,  3.57s/it, loss=0.0848, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 212/2855 [08:02<3:07:18,  4.25s/it, loss=0.0848, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 212/2855 [08:06<3:07:18,  4.25s/it, loss=0.2475, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 213/2855 [08:06<2:59:16,  4.07s/it, loss=0.2475, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 213/2855 [08:09<2:59:16,  4.07s/it, loss=0.0742, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 214/2855 [08:09<2:43:07,  3.71s/it, loss=0.0742, lr=2.87e-05]
Epoch 2/8:   7%|▋         | 214/2855 [08:11<2:43:07,  3.71s/it, loss=0.1391, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 215/2855 [08:11<2:27:17,  3.35s/it, loss=0.1391, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 215/2855 [08:18<2:27:17,  3.35s/it, loss=0.0465, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 216/2855 [08:18<3:10:26,  4.33s/it, loss=0.0465, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 216/2855 [08:23<3:10:26,  4.33s/it, loss=0.0850, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 217/2855 [08:23<3:26:00,  4.69s/it, loss=0.0850, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 217/2855 [08:27<3:26:00,  4.69s/it, loss=0.2688, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 218/2855 [08:27<3:12:20,  4.38s/it, loss=0.2688, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 218/2855 [08:30<3:12:20,  4.38s/it, loss=0.0993, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 219/2855 [08:30<2:48:14,  3.83s/it, loss=0.0993, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 219/2855 [08:32<2:48:14,  3.83s/it, loss=0.3029, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 220/2855 [08:32<2:33:43,  3.50s/it, loss=0.3029, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 220/2855 [08:35<2:33:43,  3.50s/it, loss=0.1438, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 221/2855 [08:35<2:22:51,  3.25s/it, loss=0.1438, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 221/2855 [08:38<2:22:51,  3.25s/it, loss=0.0911, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 222/2855 [08:38<2:16:31,  3.11s/it, loss=0.0911, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 222/2855 [08:40<2:16:31,  3.11s/it, loss=0.1974, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 223/2855 [08:40<2:08:11,  2.92s/it, loss=0.1974, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 223/2855 [08:43<2:08:11,  2.92s/it, loss=0.0678, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 224/2855 [08:43<2:05:01,  2.85s/it, loss=0.0678, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 224/2855 [08:46<2:05:01,  2.85s/it, loss=0.2013, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 225/2855 [08:46<2:01:31,  2.77s/it, loss=0.2013, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 225/2855 [08:48<2:01:31,  2.77s/it, loss=0.1701, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 226/2855 [08:48<2:03:49,  2.83s/it, loss=0.1701, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 226/2855 [10:36<2:03:49,  2.83s/it, loss=0.1698, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 227/2855 [10:36<24:53:55, 34.11s/it, loss=0.1698, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 227/2855 [10:41<24:53:55, 34.11s/it, loss=0.1492, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 228/2855 [10:41<18:31:37, 25.39s/it, loss=0.1492, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 228/2855 [10:44<18:31:37, 25.39s/it, loss=0.2360, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 229/2855 [10:44<13:45:32, 18.86s/it, loss=0.2360, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 229/2855 [10:46<13:45:32, 18.86s/it, loss=0.1507, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 230/2855 [10:46<10:06:53, 13.87s/it, loss=0.1507, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 230/2855 [10:50<10:06:53, 13.87s/it, loss=0.2126, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 231/2855 [10:50<7:46:52, 10.68s/it, loss=0.2126, lr=2.87e-05] 
Epoch 2/8:   8%|▊         | 231/2855 [10:52<7:46:52, 10.68s/it, loss=0.3372, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 232/2855 [10:52<6:02:47,  8.30s/it, loss=0.3372, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 232/2855 [10:55<6:02:47,  8.30s/it, loss=0.1727, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 233/2855 [10:55<4:48:16,  6.60s/it, loss=0.1727, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 233/2855 [10:58<4:48:16,  6.60s/it, loss=0.0664, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 234/2855 [10:58<3:54:08,  5.36s/it, loss=0.0664, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 234/2855 [11:00<3:54:08,  5.36s/it, loss=0.0715, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 235/2855 [11:00<3:15:53,  4.49s/it, loss=0.0715, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 235/2855 [11:03<3:15:53,  4.49s/it, loss=0.1568, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 236/2855 [11:03<2:51:49,  3.94s/it, loss=0.1568, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 236/2855 [11:05<2:51:49,  3.94s/it, loss=0.1691, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 237/2855 [11:05<2:32:31,  3.50s/it, loss=0.1691, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 237/2855 [11:08<2:32:31,  3.50s/it, loss=0.1519, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 238/2855 [11:08<2:20:46,  3.23s/it, loss=0.1519, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 238/2855 [11:11<2:20:46,  3.23s/it, loss=0.0949, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 239/2855 [11:11<2:21:32,  3.25s/it, loss=0.0949, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 239/2855 [11:14<2:21:32,  3.25s/it, loss=0.2426, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 240/2855 [11:14<2:19:44,  3.21s/it, loss=0.2426, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 240/2855 [11:17<2:19:44,  3.21s/it, loss=0.1074, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 241/2855 [11:17<2:08:56,  2.96s/it, loss=0.1074, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 241/2855 [11:20<2:08:56,  2.96s/it, loss=0.2711, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 242/2855 [11:20<2:10:07,  2.99s/it, loss=0.2711, lr=2.87e-05]
Epoch 2/8:   8%|▊         | 242/2855 [11:22<2:10:07,  2.99s/it, loss=0.0320, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 243/2855 [11:22<2:07:13,  2.92s/it, loss=0.0320, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 243/2855 [11:25<2:07:13,  2.92s/it, loss=0.0848, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 244/2855 [11:25<2:04:47,  2.87s/it, loss=0.0848, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 244/2855 [11:27<2:04:47,  2.87s/it, loss=0.1385, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 245/2855 [11:27<1:57:59,  2.71s/it, loss=0.1385, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 245/2855 [11:30<1:57:59,  2.71s/it, loss=0.2025, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 246/2855 [11:30<2:00:36,  2.77s/it, loss=0.2025, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 246/2855 [11:33<2:00:36,  2.77s/it, loss=0.1732, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 247/2855 [11:33<1:59:36,  2.75s/it, loss=0.1732, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 247/2855 [11:36<1:59:36,  2.75s/it, loss=0.0151, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 248/2855 [11:36<1:56:40,  2.69s/it, loss=0.0151, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 248/2855 [11:38<1:56:40,  2.69s/it, loss=0.0835, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 249/2855 [11:38<1:55:18,  2.65s/it, loss=0.0835, lr=2.87e-05]
Epoch 2/8:   9%|▊         | 249/2855 [11:41<1:55:18,  2.65s/it, loss=0.2373, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 250/2855 [11:41<2:02:12,  2.81s/it, loss=0.2373, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 250/2855 [11:44<2:02:12,  2.81s/it, loss=0.3181, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 251/2855 [11:44<2:05:05,  2.88s/it, loss=0.3181, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 251/2855 [11:47<2:05:05,  2.88s/it, loss=0.1939, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 252/2855 [11:47<2:01:30,  2.80s/it, loss=0.1939, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 252/2855 [11:50<2:01:30,  2.80s/it, loss=0.3266, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 253/2855 [11:50<2:04:20,  2.87s/it, loss=0.3266, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 253/2855 [11:53<2:04:20,  2.87s/it, loss=0.2102, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 254/2855 [11:53<2:03:46,  2.86s/it, loss=0.2102, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 254/2855 [11:56<2:03:46,  2.86s/it, loss=0.1752, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 255/2855 [11:56<2:01:04,  2.79s/it, loss=0.1752, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 255/2855 [11:58<2:01:04,  2.79s/it, loss=0.1401, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 256/2855 [11:58<1:56:59,  2.70s/it, loss=0.1401, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 256/2855 [12:01<1:56:59,  2.70s/it, loss=0.1431, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 257/2855 [12:01<1:57:25,  2.71s/it, loss=0.1431, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 257/2855 [12:04<1:57:25,  2.71s/it, loss=0.0574, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 258/2855 [12:04<1:58:15,  2.73s/it, loss=0.0574, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 258/2855 [12:06<1:58:15,  2.73s/it, loss=0.1300, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 259/2855 [12:06<1:56:32,  2.69s/it, loss=0.1300, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 259/2855 [12:09<1:56:32,  2.69s/it, loss=0.1194, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 260/2855 [12:09<1:55:19,  2.67s/it, loss=0.1194, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 260/2855 [12:12<1:55:19,  2.67s/it, loss=0.1520, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 261/2855 [12:12<1:58:50,  2.75s/it, loss=0.1520, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 261/2855 [12:15<1:58:50,  2.75s/it, loss=0.1544, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 262/2855 [12:15<2:01:36,  2.81s/it, loss=0.1544, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 262/2855 [12:17<2:01:36,  2.81s/it, loss=0.1644, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 263/2855 [12:17<1:56:49,  2.70s/it, loss=0.1644, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 263/2855 [12:20<1:56:49,  2.70s/it, loss=0.1555, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 264/2855 [12:20<1:58:19,  2.74s/it, loss=0.1555, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 264/2855 [12:22<1:58:19,  2.74s/it, loss=0.1680, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 265/2855 [12:22<1:54:20,  2.65s/it, loss=0.1680, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 265/2855 [12:26<1:54:20,  2.65s/it, loss=0.1578, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 266/2855 [12:26<2:01:33,  2.82s/it, loss=0.1578, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 266/2855 [12:32<2:01:33,  2.82s/it, loss=0.1390, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 267/2855 [12:32<2:46:02,  3.85s/it, loss=0.1390, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 267/2855 [12:36<2:46:02,  3.85s/it, loss=0.1987, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 268/2855 [12:36<2:46:03,  3.85s/it, loss=0.1987, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 268/2855 [12:38<2:46:03,  3.85s/it, loss=0.2134, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 269/2855 [12:38<2:28:20,  3.44s/it, loss=0.2134, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 269/2855 [12:41<2:28:20,  3.44s/it, loss=0.2207, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 270/2855 [12:41<2:17:54,  3.20s/it, loss=0.2207, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 270/2855 [12:44<2:17:54,  3.20s/it, loss=0.1025, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 271/2855 [12:44<2:13:28,  3.10s/it, loss=0.1025, lr=2.87e-05]
Epoch 2/8:   9%|▉         | 271/2855 [12:47<2:13:28,  3.10s/it, loss=0.1594, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 272/2855 [12:47<2:10:33,  3.03s/it, loss=0.1594, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 272/2855 [12:49<2:10:33,  3.03s/it, loss=0.0506, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 273/2855 [12:49<2:09:18,  3.00s/it, loss=0.0506, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 273/2855 [12:52<2:09:18,  3.00s/it, loss=0.1142, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 274/2855 [12:52<2:04:57,  2.90s/it, loss=0.1142, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 274/2855 [12:55<2:04:57,  2.90s/it, loss=0.0296, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 275/2855 [12:55<2:09:44,  3.02s/it, loss=0.0296, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 275/2855 [12:58<2:09:44,  3.02s/it, loss=0.0798, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 276/2855 [12:58<2:04:26,  2.90s/it, loss=0.0798, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 276/2855 [13:01<2:04:26,  2.90s/it, loss=0.2104, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 277/2855 [13:01<2:03:30,  2.87s/it, loss=0.2104, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 277/2855 [13:04<2:03:30,  2.87s/it, loss=0.2954, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 278/2855 [13:04<2:08:03,  2.98s/it, loss=0.2954, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 278/2855 [13:07<2:08:03,  2.98s/it, loss=0.0702, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 279/2855 [13:07<2:02:24,  2.85s/it, loss=0.0702, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 279/2855 [13:10<2:02:24,  2.85s/it, loss=0.1509, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 280/2855 [13:10<2:02:58,  2.87s/it, loss=0.1509, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 280/2855 [13:13<2:02:58,  2.87s/it, loss=0.2477, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 281/2855 [13:13<2:04:28,  2.90s/it, loss=0.2477, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 281/2855 [13:16<2:04:28,  2.90s/it, loss=0.1660, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 282/2855 [13:16<2:07:57,  2.98s/it, loss=0.1660, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 282/2855 [13:18<2:07:57,  2.98s/it, loss=0.0534, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 283/2855 [13:18<2:05:44,  2.93s/it, loss=0.0534, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 283/2855 [13:21<2:05:44,  2.93s/it, loss=0.2291, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 284/2855 [13:21<2:01:58,  2.85s/it, loss=0.2291, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 284/2855 [13:24<2:01:58,  2.85s/it, loss=0.1024, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 285/2855 [13:24<2:05:29,  2.93s/it, loss=0.1024, lr=2.87e-05]
Epoch 2/8:  10%|▉         | 285/2855 [13:27<2:05:29,  2.93s/it, loss=0.0212, lr=2.87e-05]
Epoch 2/8:  10%|█         | 286/2855 [13:27<2:00:40,  2.82s/it, loss=0.0212, lr=2.87e-05]
Epoch 2/8:  10%|█         | 286/2855 [13:29<2:00:40,  2.82s/it, loss=0.0877, lr=2.87e-05]
Epoch 2/8:  10%|█         | 287/2855 [13:29<1:57:02,  2.73s/it, loss=0.0877, lr=2.87e-05]
Epoch 2/8:  10%|█         | 287/2855 [13:32<1:57:02,  2.73s/it, loss=0.1571, lr=2.87e-05]
Epoch 2/8:  10%|█         | 288/2855 [13:32<2:01:28,  2.84s/it, loss=0.1571, lr=2.87e-05]
Epoch 2/8:  10%|█         | 288/2855 [13:35<2:01:28,  2.84s/it, loss=0.1145, lr=2.87e-05]
Epoch 2/8:  10%|█         | 289/2855 [13:35<2:02:59,  2.88s/it, loss=0.1145, lr=2.87e-05]
Epoch 2/8:  10%|█         | 289/2855 [13:38<2:02:59,  2.88s/it, loss=0.1460, lr=2.87e-05]
Epoch 2/8:  10%|█         | 290/2855 [13:38<2:00:11,  2.81s/it, loss=0.1460, lr=2.87e-05]
Epoch 2/8:  10%|█         | 290/2855 [13:41<2:00:11,  2.81s/it, loss=0.1296, lr=2.87e-05]
Epoch 2/8:  10%|█         | 291/2855 [13:41<1:56:40,  2.73s/it, loss=0.1296, lr=2.87e-05]
Epoch 2/8:  10%|█         | 291/2855 [13:44<1:56:40,  2.73s/it, loss=0.2204, lr=2.87e-05]
Epoch 2/8:  10%|█         | 292/2855 [13:44<2:01:18,  2.84s/it, loss=0.2204, lr=2.87e-05]
Epoch 2/8:  10%|█         | 292/2855 [13:46<2:01:18,  2.84s/it, loss=0.2281, lr=2.87e-05]
Epoch 2/8:  10%|█         | 293/2855 [13:46<1:57:21,  2.75s/it, loss=0.2281, lr=2.87e-05]
Epoch 2/8:  10%|█         | 293/2855 [13:49<1:57:21,  2.75s/it, loss=0.1498, lr=2.87e-05]
Epoch 2/8:  10%|█         | 294/2855 [13:49<1:57:16,  2.75s/it, loss=0.1498, lr=2.87e-05]
Epoch 2/8:  10%|█         | 294/2855 [13:52<1:57:16,  2.75s/it, loss=0.1651, lr=2.87e-05]
Epoch 2/8:  10%|█         | 295/2855 [13:52<1:59:31,  2.80s/it, loss=0.1651, lr=2.87e-05]
Epoch 2/8:  10%|█         | 295/2855 [13:55<1:59:31,  2.80s/it, loss=0.0623, lr=2.87e-05]
Epoch 2/8:  10%|█         | 296/2855 [13:55<2:00:56,  2.84s/it, loss=0.0623, lr=2.87e-05]
Epoch 2/8:  10%|█         | 296/2855 [13:57<2:00:56,  2.84s/it, loss=0.2872, lr=2.87e-05]
Epoch 2/8:  10%|█         | 297/2855 [13:57<1:56:30,  2.73s/it, loss=0.2872, lr=2.87e-05]
Epoch 2/8:  10%|█         | 297/2855 [14:00<1:56:30,  2.73s/it, loss=0.0968, lr=2.87e-05]
Epoch 2/8:  10%|█         | 298/2855 [14:00<1:56:01,  2.72s/it, loss=0.0968, lr=2.87e-05]
Epoch 2/8:  10%|█         | 298/2855 [14:03<1:56:01,  2.72s/it, loss=0.1409, lr=2.87e-05]
Epoch 2/8:  10%|█         | 299/2855 [14:03<1:57:39,  2.76s/it, loss=0.1409, lr=2.87e-05]
Epoch 2/8:  10%|█         | 299/2855 [14:06<1:57:39,  2.76s/it, loss=0.1836, lr=2.87e-05]
Epoch 2/8:  11%|█         | 300/2855 [14:06<1:58:41,  2.79s/it, loss=0.1836, lr=2.87e-05]
Epoch 2/8:  11%|█         | 300/2855 [14:08<1:58:41,  2.79s/it, loss=0.0669, lr=2.87e-05]
Epoch 2/8:  11%|█         | 301/2855 [14:08<1:53:31,  2.67s/it, loss=0.0669, lr=2.87e-05]
Epoch 2/8:  11%|█         | 301/2855 [14:10<1:53:31,  2.67s/it, loss=0.1332, lr=2.87e-05]
Epoch 2/8:  11%|█         | 302/2855 [14:10<1:48:45,  2.56s/it, loss=0.1332, lr=2.87e-05]
Epoch 2/8:  11%|█         | 302/2855 [14:14<1:48:45,  2.56s/it, loss=0.0974, lr=2.87e-05]
Epoch 2/8:  11%|█         | 303/2855 [14:14<1:56:51,  2.75s/it, loss=0.0974, lr=2.87e-05]
Epoch 2/8:  11%|█         | 303/2855 [14:16<1:56:51,  2.75s/it, loss=0.0652, lr=2.87e-05]
Epoch 2/8:  11%|█         | 304/2855 [14:16<1:56:04,  2.73s/it, loss=0.0652, lr=2.87e-05]
Epoch 2/8:  11%|█         | 304/2855 [14:19<1:56:04,  2.73s/it, loss=0.1937, lr=2.87e-05]
Epoch 2/8:  11%|█         | 305/2855 [14:19<1:54:33,  2.70s/it, loss=0.1937, lr=2.87e-05]
Epoch 2/8:  11%|█         | 305/2855 [14:22<1:54:33,  2.70s/it, loss=0.0906, lr=2.87e-05]
Epoch 2/8:  11%|█         | 306/2855 [14:22<2:01:31,  2.86s/it, loss=0.0906, lr=2.87e-05]
Epoch 2/8:  11%|█         | 306/2855 [14:25<2:01:31,  2.86s/it, loss=0.2384, lr=2.87e-05]
Epoch 2/8:  11%|█         | 307/2855 [14:25<1:58:33,  2.79s/it, loss=0.2384, lr=2.87e-05]
Epoch 2/8:  11%|█         | 307/2855 [14:27<1:58:33,  2.79s/it, loss=0.0322, lr=2.86e-05]
Epoch 2/8:  11%|█         | 308/2855 [14:27<1:56:45,  2.75s/it, loss=0.0322, lr=2.86e-05]
Epoch 2/8:  11%|█         | 308/2855 [14:30<1:56:45,  2.75s/it, loss=0.1686, lr=2.86e-05]
Epoch 2/8:  11%|█         | 309/2855 [14:30<1:52:29,  2.65s/it, loss=0.1686, lr=2.86e-05]
Epoch 2/8:  11%|█         | 309/2855 [14:33<1:52:29,  2.65s/it, loss=0.2158, lr=2.86e-05]
Epoch 2/8:  11%|█         | 310/2855 [14:33<1:56:05,  2.74s/it, loss=0.2158, lr=2.86e-05]
Epoch 2/8:  11%|█         | 310/2855 [14:36<1:56:05,  2.74s/it, loss=0.0751, lr=2.86e-05]
Epoch 2/8:  11%|█         | 311/2855 [14:36<1:56:58,  2.76s/it, loss=0.0751, lr=2.86e-05]
Epoch 2/8:  11%|█         | 311/2855 [14:38<1:56:58,  2.76s/it, loss=0.2212, lr=2.86e-05]
Epoch 2/8:  11%|█         | 312/2855 [14:38<1:52:16,  2.65s/it, loss=0.2212, lr=2.86e-05]
Epoch 2/8:  11%|█         | 312/2855 [14:40<1:52:16,  2.65s/it, loss=0.1845, lr=2.86e-05]
Epoch 2/8:  11%|█         | 313/2855 [14:40<1:50:01,  2.60s/it, loss=0.1845, lr=2.86e-05]
Epoch 2/8:  11%|█         | 313/2855 [14:43<1:50:01,  2.60s/it, loss=0.1022, lr=2.86e-05]
Epoch 2/8:  11%|█         | 314/2855 [14:43<1:50:25,  2.61s/it, loss=0.1022, lr=2.86e-05]
Epoch 2/8:  11%|█         | 314/2855 [14:46<1:50:25,  2.61s/it, loss=0.1489, lr=2.86e-05]
Epoch 2/8:  11%|█         | 315/2855 [14:46<1:50:04,  2.60s/it, loss=0.1489, lr=2.86e-05]
Epoch 2/8:  11%|█         | 315/2855 [14:49<1:50:04,  2.60s/it, loss=0.0914, lr=2.86e-05]
Epoch 2/8:  11%|█         | 316/2855 [14:49<1:54:13,  2.70s/it, loss=0.0914, lr=2.86e-05]
Epoch 2/8:  11%|█         | 316/2855 [14:51<1:54:13,  2.70s/it, loss=0.0310, lr=2.86e-05]
Epoch 2/8:  11%|█         | 317/2855 [14:51<1:53:20,  2.68s/it, loss=0.0310, lr=2.86e-05]
Epoch 2/8:  11%|█         | 317/2855 [14:54<1:53:20,  2.68s/it, loss=0.2114, lr=2.86e-05]
Epoch 2/8:  11%|█         | 318/2855 [14:54<1:56:10,  2.75s/it, loss=0.2114, lr=2.86e-05]
Epoch 2/8:  11%|█         | 318/2855 [14:57<1:56:10,  2.75s/it, loss=0.1062, lr=2.86e-05]
Epoch 2/8:  11%|█         | 319/2855 [14:57<1:58:31,  2.80s/it, loss=0.1062, lr=2.86e-05]
Epoch 2/8:  11%|█         | 319/2855 [15:00<1:58:31,  2.80s/it, loss=0.3069, lr=2.86e-05]
Epoch 2/8:  11%|█         | 320/2855 [15:00<1:54:26,  2.71s/it, loss=0.3069, lr=2.86e-05]
Epoch 2/8:  11%|█         | 320/2855 [15:02<1:54:26,  2.71s/it, loss=0.1743, lr=2.86e-05]
Epoch 2/8:  11%|█         | 321/2855 [15:02<1:52:58,  2.68s/it, loss=0.1743, lr=2.86e-05]
Epoch 2/8:  11%|█         | 321/2855 [15:05<1:52:58,  2.68s/it, loss=0.2049, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 322/2855 [15:05<1:53:55,  2.70s/it, loss=0.2049, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 322/2855 [15:08<1:53:55,  2.70s/it, loss=0.1280, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 323/2855 [15:08<1:53:15,  2.68s/it, loss=0.1280, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 323/2855 [15:10<1:53:15,  2.68s/it, loss=0.3013, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 324/2855 [15:10<1:51:39,  2.65s/it, loss=0.3013, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 324/2855 [15:13<1:51:39,  2.65s/it, loss=0.2004, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 325/2855 [15:13<1:52:33,  2.67s/it, loss=0.2004, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 325/2855 [15:16<1:52:33,  2.67s/it, loss=0.1774, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 326/2855 [15:16<1:54:28,  2.72s/it, loss=0.1774, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 326/2855 [15:18<1:54:28,  2.72s/it, loss=0.2195, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 327/2855 [15:18<1:52:26,  2.67s/it, loss=0.2195, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 327/2855 [15:21<1:52:26,  2.67s/it, loss=0.2205, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 328/2855 [15:21<1:53:08,  2.69s/it, loss=0.2205, lr=2.86e-05]
Epoch 2/8:  11%|█▏        | 328/2855 [15:24<1:53:08,  2.69s/it, loss=0.1409, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 329/2855 [15:24<1:53:22,  2.69s/it, loss=0.1409, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 329/2855 [15:27<1:53:22,  2.69s/it, loss=0.2096, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 330/2855 [15:27<1:59:39,  2.84s/it, loss=0.2096, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 330/2855 [15:29<1:59:39,  2.84s/it, loss=0.1509, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 331/2855 [15:29<1:55:36,  2.75s/it, loss=0.1509, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 331/2855 [15:32<1:55:36,  2.75s/it, loss=0.1904, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 332/2855 [15:32<1:56:37,  2.77s/it, loss=0.1904, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 332/2855 [15:36<1:56:37,  2.77s/it, loss=0.1545, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 333/2855 [15:36<2:04:10,  2.95s/it, loss=0.1545, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 333/2855 [15:38<2:04:10,  2.95s/it, loss=0.1325, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 334/2855 [15:38<1:59:30,  2.84s/it, loss=0.1325, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 334/2855 [15:41<1:59:30,  2.84s/it, loss=0.1043, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 335/2855 [15:41<2:00:14,  2.86s/it, loss=0.1043, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 335/2855 [15:44<2:00:14,  2.86s/it, loss=0.2231, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 336/2855 [15:44<1:59:07,  2.84s/it, loss=0.2231, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 336/2855 [15:47<1:59:07,  2.84s/it, loss=0.2333, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 337/2855 [15:47<1:59:03,  2.84s/it, loss=0.2333, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 337/2855 [15:49<1:59:03,  2.84s/it, loss=0.1612, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 338/2855 [15:49<1:57:49,  2.81s/it, loss=0.1612, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 338/2855 [15:52<1:57:49,  2.81s/it, loss=0.2293, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 339/2855 [15:52<1:55:28,  2.75s/it, loss=0.2293, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 339/2855 [15:55<1:55:28,  2.75s/it, loss=0.1390, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 340/2855 [15:55<1:58:25,  2.83s/it, loss=0.1390, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 340/2855 [15:58<1:58:25,  2.83s/it, loss=0.1841, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 341/2855 [15:58<1:57:38,  2.81s/it, loss=0.1841, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 341/2855 [16:01<1:57:38,  2.81s/it, loss=0.1593, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 342/2855 [16:01<2:04:20,  2.97s/it, loss=0.1593, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 342/2855 [16:04<2:04:20,  2.97s/it, loss=0.1168, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 343/2855 [16:04<1:58:57,  2.84s/it, loss=0.1168, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 343/2855 [16:06<1:58:57,  2.84s/it, loss=0.2440, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 344/2855 [16:06<1:57:49,  2.82s/it, loss=0.2440, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 344/2855 [16:09<1:57:49,  2.82s/it, loss=0.1864, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 345/2855 [16:09<1:58:58,  2.84s/it, loss=0.1864, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 345/2855 [16:54<1:58:58,  2.84s/it, loss=0.2031, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 346/2855 [16:54<10:42:47, 15.37s/it, loss=0.2031, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 346/2855 [16:56<10:42:47, 15.37s/it, loss=0.1178, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 347/2855 [16:56<7:59:23, 11.47s/it, loss=0.1178, lr=2.86e-05] 
Epoch 2/8:  12%|█▏        | 347/2855 [17:00<7:59:23, 11.47s/it, loss=0.1541, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 348/2855 [17:00<6:20:13,  9.10s/it, loss=0.1541, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 348/2855 [17:03<6:20:13,  9.10s/it, loss=0.1277, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 349/2855 [17:03<5:07:29,  7.36s/it, loss=0.1277, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 349/2855 [17:06<5:07:29,  7.36s/it, loss=0.1478, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 350/2855 [17:06<4:14:37,  6.10s/it, loss=0.1478, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 350/2855 [17:09<4:14:37,  6.10s/it, loss=0.1905, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 351/2855 [17:09<3:30:12,  5.04s/it, loss=0.1905, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 351/2855 [17:12<3:30:12,  5.04s/it, loss=0.2368, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 352/2855 [17:12<3:03:29,  4.40s/it, loss=0.2368, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 352/2855 [17:15<3:03:29,  4.40s/it, loss=0.0896, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 353/2855 [17:15<2:46:46,  4.00s/it, loss=0.0896, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 353/2855 [17:18<2:46:46,  4.00s/it, loss=0.2409, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 354/2855 [17:18<2:32:49,  3.67s/it, loss=0.2409, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 354/2855 [17:20<2:32:49,  3.67s/it, loss=0.0562, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 355/2855 [17:20<2:19:52,  3.36s/it, loss=0.0562, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 355/2855 [17:23<2:19:52,  3.36s/it, loss=0.0939, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 356/2855 [17:23<2:09:40,  3.11s/it, loss=0.0939, lr=2.86e-05]
Epoch 2/8:  12%|█▏        | 356/2855 [17:26<2:09:40,  3.11s/it, loss=0.0355, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 357/2855 [17:26<2:11:06,  3.15s/it, loss=0.0355, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 357/2855 [17:29<2:11:06,  3.15s/it, loss=0.0749, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 358/2855 [17:29<2:04:42,  3.00s/it, loss=0.0749, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 358/2855 [17:31<2:04:42,  3.00s/it, loss=0.2721, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 359/2855 [17:31<1:56:19,  2.80s/it, loss=0.2721, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 359/2855 [17:34<1:56:19,  2.80s/it, loss=0.0521, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 360/2855 [17:34<1:51:34,  2.68s/it, loss=0.0521, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 360/2855 [17:36<1:51:34,  2.68s/it, loss=0.0598, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 361/2855 [17:36<1:47:15,  2.58s/it, loss=0.0598, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 361/2855 [17:38<1:47:15,  2.58s/it, loss=0.1625, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 362/2855 [17:38<1:42:55,  2.48s/it, loss=0.1625, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 362/2855 [17:40<1:42:55,  2.48s/it, loss=0.1254, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 363/2855 [17:40<1:38:58,  2.38s/it, loss=0.1254, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 363/2855 [17:42<1:38:58,  2.38s/it, loss=0.0760, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 364/2855 [17:42<1:35:50,  2.31s/it, loss=0.0760, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 364/2855 [17:45<1:35:50,  2.31s/it, loss=0.0214, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 365/2855 [17:45<1:34:31,  2.28s/it, loss=0.0214, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 365/2855 [17:47<1:34:31,  2.28s/it, loss=0.1155, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 366/2855 [17:47<1:37:12,  2.34s/it, loss=0.1155, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 366/2855 [17:49<1:37:12,  2.34s/it, loss=0.1984, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 367/2855 [17:49<1:34:57,  2.29s/it, loss=0.1984, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 367/2855 [17:52<1:34:57,  2.29s/it, loss=0.0498, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 368/2855 [17:52<1:33:27,  2.25s/it, loss=0.0498, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 368/2855 [17:54<1:33:27,  2.25s/it, loss=0.0260, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 369/2855 [17:54<1:36:24,  2.33s/it, loss=0.0260, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 369/2855 [17:56<1:36:24,  2.33s/it, loss=0.1519, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 370/2855 [17:56<1:37:35,  2.36s/it, loss=0.1519, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 370/2855 [17:59<1:37:35,  2.36s/it, loss=0.1727, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 371/2855 [17:59<1:36:51,  2.34s/it, loss=0.1727, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 371/2855 [18:01<1:36:51,  2.34s/it, loss=0.0971, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 372/2855 [18:01<1:33:49,  2.27s/it, loss=0.0971, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 372/2855 [18:03<1:33:49,  2.27s/it, loss=0.1600, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 373/2855 [18:03<1:32:20,  2.23s/it, loss=0.1600, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 373/2855 [18:05<1:32:20,  2.23s/it, loss=0.2542, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 374/2855 [18:05<1:31:42,  2.22s/it, loss=0.2542, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 374/2855 [18:08<1:31:42,  2.22s/it, loss=0.2709, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 375/2855 [18:08<1:33:52,  2.27s/it, loss=0.2709, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 375/2855 [18:10<1:33:52,  2.27s/it, loss=0.1555, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 376/2855 [18:10<1:31:33,  2.22s/it, loss=0.1555, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 376/2855 [18:12<1:31:33,  2.22s/it, loss=0.0280, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 377/2855 [18:12<1:30:20,  2.19s/it, loss=0.0280, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 377/2855 [18:14<1:30:20,  2.19s/it, loss=0.2370, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 378/2855 [18:14<1:32:32,  2.24s/it, loss=0.2370, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 378/2855 [18:17<1:32:32,  2.24s/it, loss=0.1076, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 379/2855 [18:17<1:34:12,  2.28s/it, loss=0.1076, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 379/2855 [18:19<1:34:12,  2.28s/it, loss=0.2726, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 380/2855 [18:19<1:33:21,  2.26s/it, loss=0.2726, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 380/2855 [18:21<1:33:21,  2.26s/it, loss=0.0990, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 381/2855 [18:21<1:29:16,  2.17s/it, loss=0.0990, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 381/2855 [18:23<1:29:16,  2.17s/it, loss=0.1684, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 382/2855 [18:23<1:26:18,  2.09s/it, loss=0.1684, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 382/2855 [18:25<1:26:18,  2.09s/it, loss=0.1919, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 383/2855 [18:25<1:23:43,  2.03s/it, loss=0.1919, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 383/2855 [18:26<1:23:43,  2.03s/it, loss=0.1966, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 384/2855 [18:26<1:22:23,  2.00s/it, loss=0.1966, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 384/2855 [18:29<1:22:23,  2.00s/it, loss=0.1615, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 385/2855 [18:29<1:33:58,  2.28s/it, loss=0.1615, lr=2.86e-05]
Epoch 2/8:  13%|█▎        | 385/2855 [18:31<1:33:58,  2.28s/it, loss=0.1140, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 386/2855 [18:31<1:28:45,  2.16s/it, loss=0.1140, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 386/2855 [18:34<1:28:45,  2.16s/it, loss=0.2021, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 387/2855 [18:34<1:31:08,  2.22s/it, loss=0.2021, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 387/2855 [18:36<1:31:08,  2.22s/it, loss=0.2407, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 388/2855 [18:36<1:29:38,  2.18s/it, loss=0.2407, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 388/2855 [18:38<1:29:38,  2.18s/it, loss=0.0843, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 389/2855 [18:38<1:30:49,  2.21s/it, loss=0.0843, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 389/2855 [18:40<1:30:49,  2.21s/it, loss=0.2949, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 390/2855 [18:40<1:29:00,  2.17s/it, loss=0.2949, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 390/2855 [18:42<1:29:00,  2.17s/it, loss=0.1324, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 391/2855 [18:42<1:28:10,  2.15s/it, loss=0.1324, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 391/2855 [18:44<1:28:10,  2.15s/it, loss=0.1676, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 392/2855 [18:44<1:29:54,  2.19s/it, loss=0.1676, lr=2.86e-05]
Epoch 2/8:  14%|█▎        | 392/2855 [18:47<1:29:54,  2.19s/it, loss=0.1189, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 393/2855 [18:47<1:29:12,  2.17s/it, loss=0.1189, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 393/2855 [18:49<1:29:12,  2.17s/it, loss=0.1865, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 394/2855 [18:49<1:31:05,  2.22s/it, loss=0.1865, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 394/2855 [18:51<1:31:05,  2.22s/it, loss=0.1003, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 395/2855 [18:51<1:27:41,  2.14s/it, loss=0.1003, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 395/2855 [18:53<1:27:41,  2.14s/it, loss=0.2565, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 396/2855 [18:53<1:26:04,  2.10s/it, loss=0.2565, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 396/2855 [18:55<1:26:04,  2.10s/it, loss=0.1399, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 397/2855 [18:55<1:23:21,  2.03s/it, loss=0.1399, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 397/2855 [18:57<1:23:21,  2.03s/it, loss=0.0906, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 398/2855 [18:57<1:21:50,  2.00s/it, loss=0.0906, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 398/2855 [18:59<1:21:50,  2.00s/it, loss=0.1594, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 399/2855 [18:59<1:28:55,  2.17s/it, loss=0.1594, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 399/2855 [19:01<1:28:55,  2.17s/it, loss=0.1495, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 400/2855 [19:01<1:25:14,  2.08s/it, loss=0.1495, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 400/2855 [19:03<1:25:14,  2.08s/it, loss=0.1219, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 401/2855 [19:03<1:24:10,  2.06s/it, loss=0.1219, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 401/2855 [19:05<1:24:10,  2.06s/it, loss=0.0423, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 402/2855 [19:05<1:26:37,  2.12s/it, loss=0.0423, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 402/2855 [19:08<1:26:37,  2.12s/it, loss=0.2290, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 403/2855 [19:08<1:27:46,  2.15s/it, loss=0.2290, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 403/2855 [19:10<1:27:46,  2.15s/it, loss=0.1552, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 404/2855 [19:10<1:26:38,  2.12s/it, loss=0.1552, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 404/2855 [19:12<1:26:38,  2.12s/it, loss=0.0517, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 405/2855 [19:12<1:23:40,  2.05s/it, loss=0.0517, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 405/2855 [19:14<1:23:40,  2.05s/it, loss=0.2351, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 406/2855 [19:14<1:24:40,  2.07s/it, loss=0.2351, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 406/2855 [19:16<1:24:40,  2.07s/it, loss=0.0817, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 407/2855 [19:16<1:24:48,  2.08s/it, loss=0.0817, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 407/2855 [19:18<1:24:48,  2.08s/it, loss=0.3003, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 408/2855 [19:18<1:26:28,  2.12s/it, loss=0.3003, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 408/2855 [19:20<1:26:28,  2.12s/it, loss=0.2947, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 409/2855 [19:20<1:25:04,  2.09s/it, loss=0.2947, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 409/2855 [19:22<1:25:04,  2.09s/it, loss=0.1866, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 410/2855 [19:22<1:21:47,  2.01s/it, loss=0.1866, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 410/2855 [19:24<1:21:47,  2.01s/it, loss=0.0844, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 411/2855 [19:24<1:20:18,  1.97s/it, loss=0.0844, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 411/2855 [19:25<1:20:18,  1.97s/it, loss=0.1842, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 412/2855 [19:25<1:17:48,  1.91s/it, loss=0.1842, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 412/2855 [19:27<1:17:48,  1.91s/it, loss=0.1757, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 413/2855 [19:27<1:17:57,  1.92s/it, loss=0.1757, lr=2.86e-05]
Epoch 2/8:  14%|█▍        | 413/2855 [19:30<1:17:57,  1.92s/it, loss=0.1571, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 414/2855 [19:30<1:21:50,  2.01s/it, loss=0.1571, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 414/2855 [19:32<1:21:50,  2.01s/it, loss=0.0973, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 415/2855 [19:32<1:22:08,  2.02s/it, loss=0.0973, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 415/2855 [19:34<1:22:08,  2.02s/it, loss=0.0767, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 416/2855 [19:34<1:23:09,  2.05s/it, loss=0.0767, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 416/2855 [19:36<1:23:09,  2.05s/it, loss=0.1919, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 417/2855 [19:36<1:23:29,  2.05s/it, loss=0.1919, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 417/2855 [19:38<1:23:29,  2.05s/it, loss=0.1519, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 418/2855 [19:38<1:22:51,  2.04s/it, loss=0.1519, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 418/2855 [19:40<1:22:51,  2.04s/it, loss=0.2135, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 419/2855 [19:40<1:22:56,  2.04s/it, loss=0.2135, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 419/2855 [19:42<1:22:56,  2.04s/it, loss=0.1138, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 420/2855 [19:42<1:20:28,  1.98s/it, loss=0.1138, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 420/2855 [19:44<1:20:28,  1.98s/it, loss=0.0764, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 421/2855 [19:44<1:21:26,  2.01s/it, loss=0.0764, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 421/2855 [19:46<1:21:26,  2.01s/it, loss=0.2422, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 422/2855 [19:46<1:21:37,  2.01s/it, loss=0.2422, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 422/2855 [19:48<1:21:37,  2.01s/it, loss=0.1321, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 423/2855 [19:48<1:22:51,  2.04s/it, loss=0.1321, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 423/2855 [19:50<1:22:51,  2.04s/it, loss=0.1047, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 424/2855 [19:50<1:23:13,  2.05s/it, loss=0.1047, lr=2.86e-05]
Epoch 2/8:  15%|█▍        | 424/2855 [19:52<1:23:13,  2.05s/it, loss=0.1171, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 425/2855 [19:52<1:20:59,  2.00s/it, loss=0.1171, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 425/2855 [19:54<1:20:59,  2.00s/it, loss=0.2453, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 426/2855 [19:54<1:19:10,  1.96s/it, loss=0.2453, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 426/2855 [19:56<1:19:10,  1.96s/it, loss=0.2492, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 427/2855 [19:56<1:18:09,  1.93s/it, loss=0.2492, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 427/2855 [19:58<1:18:09,  1.93s/it, loss=0.1724, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 428/2855 [19:58<1:17:40,  1.92s/it, loss=0.1724, lr=2.85e-05]
Epoch 2/8:  15%|█▍        | 428/2855 [20:00<1:17:40,  1.92s/it, loss=0.0857, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 429/2855 [20:00<1:20:35,  1.99s/it, loss=0.0857, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 429/2855 [20:02<1:20:35,  1.99s/it, loss=0.0722, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 430/2855 [20:02<1:19:54,  1.98s/it, loss=0.0722, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 430/2855 [20:03<1:19:54,  1.98s/it, loss=0.1955, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 431/2855 [20:03<1:17:16,  1.91s/it, loss=0.1955, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 431/2855 [20:05<1:17:16,  1.91s/it, loss=0.1609, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 432/2855 [20:05<1:14:40,  1.85s/it, loss=0.1609, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 432/2855 [20:07<1:14:40,  1.85s/it, loss=0.1441, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 433/2855 [20:07<1:12:56,  1.81s/it, loss=0.1441, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 433/2855 [20:09<1:12:56,  1.81s/it, loss=0.2066, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 434/2855 [20:09<1:13:01,  1.81s/it, loss=0.2066, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 434/2855 [20:10<1:13:01,  1.81s/it, loss=0.0362, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 435/2855 [20:10<1:11:02,  1.76s/it, loss=0.0362, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 435/2855 [20:12<1:11:02,  1.76s/it, loss=0.1963, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 436/2855 [20:12<1:07:39,  1.68s/it, loss=0.1963, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 436/2855 [20:14<1:07:39,  1.68s/it, loss=0.1124, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 437/2855 [20:14<1:09:45,  1.73s/it, loss=0.1124, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 437/2855 [20:16<1:09:45,  1.73s/it, loss=0.2724, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 438/2855 [20:16<1:15:04,  1.86s/it, loss=0.2724, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 438/2855 [20:18<1:15:04,  1.86s/it, loss=0.0794, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 439/2855 [20:18<1:17:08,  1.92s/it, loss=0.0794, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 439/2855 [20:20<1:17:08,  1.92s/it, loss=0.1860, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 440/2855 [20:20<1:17:46,  1.93s/it, loss=0.1860, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 440/2855 [20:21<1:17:46,  1.93s/it, loss=0.1215, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 441/2855 [20:21<1:15:13,  1.87s/it, loss=0.1215, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 441/2855 [20:23<1:15:13,  1.87s/it, loss=0.0715, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 442/2855 [20:23<1:14:01,  1.84s/it, loss=0.0715, lr=2.85e-05]
Epoch 2/8:  15%|█▌        | 442/2855 [20:25<1:14:01,  1.84s/it, loss=0.0720, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 443/2855 [20:25<1:14:53,  1.86s/it, loss=0.0720, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 443/2855 [20:27<1:14:53,  1.86s/it, loss=0.1473, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 444/2855 [20:27<1:15:04,  1.87s/it, loss=0.1473, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 444/2855 [20:29<1:15:04,  1.87s/it, loss=0.1345, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 445/2855 [20:29<1:17:56,  1.94s/it, loss=0.1345, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 445/2855 [20:31<1:17:56,  1.94s/it, loss=0.1645, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 446/2855 [20:31<1:16:05,  1.90s/it, loss=0.1645, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 446/2855 [20:33<1:16:05,  1.90s/it, loss=0.0445, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 447/2855 [20:33<1:13:41,  1.84s/it, loss=0.0445, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 447/2855 [20:35<1:13:41,  1.84s/it, loss=0.0383, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 448/2855 [20:35<1:14:42,  1.86s/it, loss=0.0383, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 448/2855 [20:36<1:14:42,  1.86s/it, loss=0.2237, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 449/2855 [20:36<1:14:39,  1.86s/it, loss=0.2237, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 449/2855 [20:38<1:14:39,  1.86s/it, loss=0.0988, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 450/2855 [20:38<1:13:50,  1.84s/it, loss=0.0988, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 450/2855 [20:40<1:13:50,  1.84s/it, loss=0.1418, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 451/2855 [20:40<1:12:56,  1.82s/it, loss=0.1418, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 451/2855 [20:41<1:12:56,  1.82s/it, loss=0.1655, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 452/2855 [20:41<1:09:00,  1.72s/it, loss=0.1655, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 452/2855 [20:43<1:09:00,  1.72s/it, loss=0.1667, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 453/2855 [20:43<1:07:07,  1.68s/it, loss=0.1667, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 453/2855 [20:45<1:07:07,  1.68s/it, loss=0.2101, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 454/2855 [20:45<1:06:37,  1.66s/it, loss=0.2101, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 454/2855 [20:46<1:06:37,  1.66s/it, loss=0.0948, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 455/2855 [20:46<1:06:13,  1.66s/it, loss=0.0948, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 455/2855 [20:48<1:06:13,  1.66s/it, loss=0.1659, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 456/2855 [20:48<1:11:55,  1.80s/it, loss=0.1659, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 456/2855 [20:50<1:11:55,  1.80s/it, loss=0.1166, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 457/2855 [20:50<1:13:20,  1.83s/it, loss=0.1166, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 457/2855 [20:52<1:13:20,  1.83s/it, loss=0.1449, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 458/2855 [20:52<1:10:53,  1.77s/it, loss=0.1449, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 458/2855 [20:54<1:10:53,  1.77s/it, loss=0.1849, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 459/2855 [20:54<1:10:29,  1.77s/it, loss=0.1849, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 459/2855 [20:56<1:10:29,  1.77s/it, loss=0.1547, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 460/2855 [20:56<1:11:33,  1.79s/it, loss=0.1547, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 460/2855 [20:57<1:11:33,  1.79s/it, loss=0.2036, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 461/2855 [20:57<1:11:48,  1.80s/it, loss=0.2036, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 461/2855 [20:59<1:11:48,  1.80s/it, loss=0.1738, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 462/2855 [20:59<1:13:59,  1.86s/it, loss=0.1738, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 462/2855 [21:01<1:13:59,  1.86s/it, loss=0.1822, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 463/2855 [21:01<1:12:04,  1.81s/it, loss=0.1822, lr=2.85e-05]
Epoch 2/8:  16%|█▌        | 463/2855 [21:03<1:12:04,  1.81s/it, loss=0.1801, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 464/2855 [21:03<1:10:15,  1.76s/it, loss=0.1801, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 464/2855 [21:05<1:10:15,  1.76s/it, loss=0.1169, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 465/2855 [21:05<1:12:27,  1.82s/it, loss=0.1169, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 465/2855 [21:07<1:12:27,  1.82s/it, loss=0.1615, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 466/2855 [21:07<1:12:03,  1.81s/it, loss=0.1615, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 466/2855 [21:08<1:12:03,  1.81s/it, loss=0.1197, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 467/2855 [21:08<1:10:48,  1.78s/it, loss=0.1197, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 467/2855 [21:10<1:10:48,  1.78s/it, loss=0.1256, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 468/2855 [21:10<1:09:53,  1.76s/it, loss=0.1256, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 468/2855 [21:11<1:09:53,  1.76s/it, loss=0.2127, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 469/2855 [21:11<1:06:18,  1.67s/it, loss=0.2127, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 469/2855 [21:13<1:06:18,  1.67s/it, loss=0.1825, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 470/2855 [21:13<1:05:02,  1.64s/it, loss=0.1825, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 470/2855 [21:15<1:05:02,  1.64s/it, loss=0.1522, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 471/2855 [21:15<1:08:52,  1.73s/it, loss=0.1522, lr=2.85e-05]
Epoch 2/8:  16%|█▋        | 471/2855 [21:17<1:08:52,  1.73s/it, loss=0.0746, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 472/2855 [21:17<1:10:37,  1.78s/it, loss=0.0746, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 472/2855 [21:19<1:10:37,  1.78s/it, loss=0.0608, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 473/2855 [21:19<1:10:50,  1.78s/it, loss=0.0608, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 473/2855 [21:20<1:10:50,  1.78s/it, loss=0.2386, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 474/2855 [21:20<1:11:09,  1.79s/it, loss=0.2386, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 474/2855 [21:22<1:11:09,  1.79s/it, loss=0.2307, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 475/2855 [21:22<1:09:28,  1.75s/it, loss=0.2307, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 475/2855 [21:24<1:09:28,  1.75s/it, loss=0.2478, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 476/2855 [21:24<1:09:47,  1.76s/it, loss=0.2478, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 476/2855 [21:26<1:09:47,  1.76s/it, loss=0.2374, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 477/2855 [21:26<1:10:47,  1.79s/it, loss=0.2374, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 477/2855 [21:27<1:10:47,  1.79s/it, loss=0.0885, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 478/2855 [21:27<1:10:24,  1.78s/it, loss=0.0885, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 478/2855 [21:29<1:10:24,  1.78s/it, loss=0.1106, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 479/2855 [21:29<1:11:29,  1.81s/it, loss=0.1106, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 479/2855 [21:31<1:11:29,  1.81s/it, loss=0.1259, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 480/2855 [21:31<1:10:24,  1.78s/it, loss=0.1259, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 480/2855 [21:33<1:10:24,  1.78s/it, loss=0.0669, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 481/2855 [21:33<1:08:40,  1.74s/it, loss=0.0669, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 481/2855 [21:34<1:08:40,  1.74s/it, loss=0.0933, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 482/2855 [21:34<1:08:21,  1.73s/it, loss=0.0933, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 482/2855 [21:36<1:08:21,  1.73s/it, loss=0.1030, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 483/2855 [21:36<1:06:33,  1.68s/it, loss=0.1030, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 483/2855 [21:37<1:06:33,  1.68s/it, loss=0.0436, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 484/2855 [21:37<1:03:45,  1.61s/it, loss=0.0436, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 484/2855 [21:39<1:03:45,  1.61s/it, loss=0.1007, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 485/2855 [21:39<1:03:31,  1.61s/it, loss=0.1007, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 485/2855 [21:41<1:03:31,  1.61s/it, loss=0.1902, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 486/2855 [21:41<1:04:44,  1.64s/it, loss=0.1902, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 486/2855 [21:42<1:04:44,  1.64s/it, loss=0.2343, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 487/2855 [21:42<1:03:20,  1.61s/it, loss=0.2343, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 487/2855 [21:44<1:03:20,  1.61s/it, loss=0.3593, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 488/2855 [21:44<1:02:08,  1.58s/it, loss=0.3593, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 488/2855 [21:45<1:02:08,  1.58s/it, loss=0.1990, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 489/2855 [21:45<1:01:34,  1.56s/it, loss=0.1990, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 489/2855 [21:47<1:01:34,  1.56s/it, loss=0.1237, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 490/2855 [21:47<1:01:08,  1.55s/it, loss=0.1237, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 490/2855 [21:49<1:01:08,  1.55s/it, loss=0.0359, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 491/2855 [21:49<1:05:48,  1.67s/it, loss=0.0359, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 491/2855 [21:51<1:05:48,  1.67s/it, loss=0.1542, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 492/2855 [21:51<1:08:25,  1.74s/it, loss=0.1542, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 492/2855 [21:52<1:08:25,  1.74s/it, loss=0.0378, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 493/2855 [21:52<1:06:10,  1.68s/it, loss=0.0378, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 493/2855 [21:54<1:06:10,  1.68s/it, loss=0.1879, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 494/2855 [21:54<1:05:47,  1.67s/it, loss=0.1879, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 494/2855 [21:56<1:05:47,  1.67s/it, loss=0.2239, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 495/2855 [21:56<1:06:56,  1.70s/it, loss=0.2239, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 495/2855 [21:57<1:06:56,  1.70s/it, loss=0.1604, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 496/2855 [21:57<1:07:40,  1.72s/it, loss=0.1604, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 496/2855 [21:59<1:07:40,  1.72s/it, loss=0.1511, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 497/2855 [21:59<1:09:50,  1.78s/it, loss=0.1511, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 497/2855 [22:01<1:09:50,  1.78s/it, loss=0.1473, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 498/2855 [22:01<1:09:47,  1.78s/it, loss=0.1473, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 498/2855 [22:03<1:09:47,  1.78s/it, loss=0.1399, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 499/2855 [22:03<1:07:47,  1.73s/it, loss=0.1399, lr=2.85e-05]
Epoch 2/8:  17%|█▋        | 499/2855 [22:04<1:07:47,  1.73s/it, loss=0.1203, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 500/2855 [22:04<1:07:42,  1.72s/it, loss=0.1203, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 500/2855 [22:06<1:07:42,  1.72s/it, loss=0.1169, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 501/2855 [22:06<1:08:40,  1.75s/it, loss=0.1169, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 501/2855 [22:08<1:08:40,  1.75s/it, loss=0.2016, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 502/2855 [22:08<1:06:58,  1.71s/it, loss=0.2016, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 502/2855 [22:09<1:06:58,  1.71s/it, loss=0.1831, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 503/2855 [22:09<1:05:44,  1.68s/it, loss=0.1831, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 503/2855 [22:11<1:05:44,  1.68s/it, loss=0.0988, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 504/2855 [22:11<1:04:51,  1.66s/it, loss=0.0988, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 504/2855 [22:12<1:04:51,  1.66s/it, loss=0.1831, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 505/2855 [22:12<1:02:33,  1.60s/it, loss=0.1831, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 505/2855 [22:14<1:02:33,  1.60s/it, loss=0.1376, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 506/2855 [22:14<1:01:32,  1.57s/it, loss=0.1376, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 506/2855 [22:16<1:01:32,  1.57s/it, loss=0.1303, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 507/2855 [22:16<1:01:27,  1.57s/it, loss=0.1303, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 507/2855 [22:17<1:01:27,  1.57s/it, loss=0.0393, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 508/2855 [22:17<1:00:18,  1.54s/it, loss=0.0393, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 508/2855 [22:19<1:00:18,  1.54s/it, loss=0.1509, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 509/2855 [22:19<1:00:59,  1.56s/it, loss=0.1509, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 509/2855 [22:20<1:00:59,  1.56s/it, loss=0.0704, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 510/2855 [22:20<1:03:32,  1.63s/it, loss=0.0704, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 510/2855 [22:22<1:03:32,  1.63s/it, loss=0.2519, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 511/2855 [22:22<1:05:02,  1.67s/it, loss=0.2519, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 511/2855 [22:24<1:05:02,  1.67s/it, loss=0.0815, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 512/2855 [22:24<1:05:34,  1.68s/it, loss=0.0815, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 512/2855 [22:26<1:05:34,  1.68s/it, loss=0.1454, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 513/2855 [22:26<1:06:17,  1.70s/it, loss=0.1454, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 513/2855 [22:27<1:06:17,  1.70s/it, loss=0.2084, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 514/2855 [22:27<1:06:34,  1.71s/it, loss=0.2084, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 514/2855 [22:29<1:06:34,  1.71s/it, loss=0.2209, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 515/2855 [22:29<1:07:18,  1.73s/it, loss=0.2209, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 515/2855 [22:31<1:07:18,  1.73s/it, loss=0.1555, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 516/2855 [22:31<1:08:31,  1.76s/it, loss=0.1555, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 516/2855 [22:33<1:08:31,  1.76s/it, loss=0.2112, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 517/2855 [22:33<1:06:47,  1.71s/it, loss=0.2112, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 517/2855 [22:34<1:06:47,  1.71s/it, loss=0.1771, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 518/2855 [22:34<1:06:30,  1.71s/it, loss=0.1771, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 518/2855 [22:36<1:06:30,  1.71s/it, loss=0.1359, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 519/2855 [22:36<1:08:25,  1.76s/it, loss=0.1359, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 519/2855 [22:38<1:08:25,  1.76s/it, loss=0.0664, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 520/2855 [22:38<1:07:42,  1.74s/it, loss=0.0664, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 520/2855 [22:40<1:07:42,  1.74s/it, loss=0.0983, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 521/2855 [22:40<1:08:18,  1.76s/it, loss=0.0983, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 521/2855 [22:41<1:08:18,  1.76s/it, loss=0.0893, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 522/2855 [22:41<1:08:37,  1.76s/it, loss=0.0893, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 522/2855 [22:43<1:08:37,  1.76s/it, loss=0.1838, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 523/2855 [22:43<1:04:41,  1.66s/it, loss=0.1838, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 523/2855 [22:44<1:04:41,  1.66s/it, loss=0.0679, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 524/2855 [22:44<1:02:26,  1.61s/it, loss=0.0679, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 524/2855 [22:46<1:02:26,  1.61s/it, loss=0.1336, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 525/2855 [22:46<1:01:40,  1.59s/it, loss=0.1336, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 525/2855 [22:47<1:01:40,  1.59s/it, loss=0.1190, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 526/2855 [22:47<1:01:24,  1.58s/it, loss=0.1190, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 526/2855 [22:49<1:01:24,  1.58s/it, loss=0.1929, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 527/2855 [22:49<1:04:30,  1.66s/it, loss=0.1929, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 527/2855 [22:51<1:04:30,  1.66s/it, loss=0.2146, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 528/2855 [22:51<1:07:06,  1.73s/it, loss=0.2146, lr=2.85e-05]
Epoch 2/8:  18%|█▊        | 528/2855 [22:53<1:07:06,  1.73s/it, loss=0.0552, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 529/2855 [22:53<1:05:02,  1.68s/it, loss=0.0552, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 529/2855 [22:54<1:05:02,  1.68s/it, loss=0.0741, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 530/2855 [22:54<1:03:50,  1.65s/it, loss=0.0741, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 530/2855 [22:56<1:03:50,  1.65s/it, loss=0.1718, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 531/2855 [22:56<1:04:46,  1.67s/it, loss=0.1718, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 531/2855 [22:58<1:04:46,  1.67s/it, loss=0.1178, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 532/2855 [22:58<1:04:59,  1.68s/it, loss=0.1178, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 532/2855 [23:00<1:04:59,  1.68s/it, loss=0.1238, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 533/2855 [23:00<1:06:07,  1.71s/it, loss=0.1238, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 533/2855 [23:01<1:06:07,  1.71s/it, loss=0.1519, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 534/2855 [23:01<1:07:03,  1.73s/it, loss=0.1519, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 534/2855 [23:03<1:07:03,  1.73s/it, loss=0.1050, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 535/2855 [23:03<1:04:54,  1.68s/it, loss=0.1050, lr=2.85e-05]
Epoch 2/8:  19%|█▊        | 535/2855 [23:04<1:04:54,  1.68s/it, loss=0.1092, lr=2.85e-05]
Epoch 2/8:  19%|█▉        | 536/2855 [23:04<1:03:37,  1.65s/it, loss=0.1092, lr=2.85e-05]
Epoch 2/8:  19%|█▉        | 536/2855 [23:06<1:03:37,  1.65s/it, loss=0.0648, lr=2.85e-05]
Epoch 2/8:  19%|█▉        | 537/2855 [23:06<1:05:40,  1.70s/it, loss=0.0648, lr=2.85e-05]
Epoch 2/8:  19%|█▉        | 537/2855 [23:08<1:05:40,  1.70s/it, loss=0.3034, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 538/2855 [23:08<1:04:47,  1.68s/it, loss=0.3034, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 538/2855 [23:09<1:04:47,  1.68s/it, loss=0.2239, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 539/2855 [23:09<1:02:23,  1.62s/it, loss=0.2239, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 539/2855 [23:11<1:02:23,  1.62s/it, loss=0.0874, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 540/2855 [23:11<1:01:40,  1.60s/it, loss=0.0874, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 540/2855 [23:12<1:01:40,  1.60s/it, loss=0.0752, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 541/2855 [23:12<59:03,  1.53s/it, loss=0.0752, lr=2.84e-05]  
Epoch 2/8:  19%|█▉        | 541/2855 [23:14<59:03,  1.53s/it, loss=0.2297, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 542/2855 [23:14<57:16,  1.49s/it, loss=0.2297, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 542/2855 [23:15<57:16,  1.49s/it, loss=0.1037, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 543/2855 [23:15<58:18,  1.51s/it, loss=0.1037, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 543/2855 [23:17<58:18,  1.51s/it, loss=0.1679, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 544/2855 [23:17<1:01:33,  1.60s/it, loss=0.1679, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 544/2855 [23:19<1:01:33,  1.60s/it, loss=0.1550, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 545/2855 [23:19<1:01:18,  1.59s/it, loss=0.1550, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 545/2855 [23:20<1:01:18,  1.59s/it, loss=0.2429, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 546/2855 [23:20<1:02:24,  1.62s/it, loss=0.2429, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 546/2855 [23:22<1:02:24,  1.62s/it, loss=0.1364, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 547/2855 [23:22<1:03:19,  1.65s/it, loss=0.1364, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 547/2855 [23:24<1:03:19,  1.65s/it, loss=0.2917, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 548/2855 [23:24<1:02:52,  1.64s/it, loss=0.2917, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 548/2855 [23:25<1:02:52,  1.64s/it, loss=0.0716, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 549/2855 [23:25<1:04:25,  1.68s/it, loss=0.0716, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 549/2855 [23:27<1:04:25,  1.68s/it, loss=0.0645, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 550/2855 [23:27<1:05:42,  1.71s/it, loss=0.0645, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 550/2855 [23:29<1:05:42,  1.71s/it, loss=0.0664, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 551/2855 [23:29<1:04:51,  1.69s/it, loss=0.0664, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 551/2855 [23:31<1:04:51,  1.69s/it, loss=0.1003, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 552/2855 [23:31<1:05:37,  1.71s/it, loss=0.1003, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 552/2855 [23:32<1:05:37,  1.71s/it, loss=0.2378, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 553/2855 [23:32<1:05:07,  1.70s/it, loss=0.2378, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 553/2855 [23:34<1:05:07,  1.70s/it, loss=0.1576, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 554/2855 [23:34<1:03:19,  1.65s/it, loss=0.1576, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 554/2855 [23:35<1:03:19,  1.65s/it, loss=0.0921, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 555/2855 [23:35<1:03:35,  1.66s/it, loss=0.0921, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 555/2855 [23:37<1:03:35,  1.66s/it, loss=0.1700, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 556/2855 [23:37<1:04:27,  1.68s/it, loss=0.1700, lr=2.84e-05]
Epoch 2/8:  19%|█▉        | 556/2855 [23:39<1:04:27,  1.68s/it, loss=0.1789, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 557/2855 [23:39<1:04:25,  1.68s/it, loss=0.1789, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 557/2855 [23:41<1:04:25,  1.68s/it, loss=0.1163, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 558/2855 [23:41<1:06:08,  1.73s/it, loss=0.1163, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 558/2855 [23:42<1:06:08,  1.73s/it, loss=0.1488, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 559/2855 [23:42<1:05:46,  1.72s/it, loss=0.1488, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 559/2855 [23:44<1:05:46,  1.72s/it, loss=0.0055, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 560/2855 [23:44<1:01:28,  1.61s/it, loss=0.0055, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 560/2855 [23:45<1:01:28,  1.61s/it, loss=0.2195, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 561/2855 [23:45<59:42,  1.56s/it, loss=0.2195, lr=2.84e-05]  
Epoch 2/8:  20%|█▉        | 561/2855 [23:47<59:42,  1.56s/it, loss=0.1206, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 562/2855 [23:47<58:55,  1.54s/it, loss=0.1206, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 562/2855 [23:48<58:55,  1.54s/it, loss=0.1533, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 563/2855 [23:48<1:00:19,  1.58s/it, loss=0.1533, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 563/2855 [23:50<1:00:19,  1.58s/it, loss=0.1865, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 564/2855 [23:50<1:00:55,  1.60s/it, loss=0.1865, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 564/2855 [23:52<1:00:55,  1.60s/it, loss=0.1392, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 565/2855 [23:52<1:01:25,  1.61s/it, loss=0.1392, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 565/2855 [23:53<1:01:25,  1.61s/it, loss=0.2492, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 566/2855 [23:53<58:13,  1.53s/it, loss=0.2492, lr=2.84e-05]  
Epoch 2/8:  20%|█▉        | 566/2855 [23:54<58:13,  1.53s/it, loss=0.1079, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 567/2855 [23:54<55:47,  1.46s/it, loss=0.1079, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 567/2855 [23:56<55:47,  1.46s/it, loss=0.2995, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 568/2855 [23:56<55:36,  1.46s/it, loss=0.2995, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 568/2855 [23:57<55:36,  1.46s/it, loss=0.0606, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 569/2855 [23:57<55:28,  1.46s/it, loss=0.0606, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 569/2855 [23:59<55:28,  1.46s/it, loss=0.0600, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 570/2855 [23:59<57:32,  1.51s/it, loss=0.0600, lr=2.84e-05]
Epoch 2/8:  20%|█▉        | 570/2855 [24:01<57:32,  1.51s/it, loss=0.0541, lr=2.84e-05]
Epoch 2/8:  20%|██        | 571/2855 [24:01<1:01:05,  1.61s/it, loss=0.0541, lr=2.84e-05]
Epoch 2/8:  20%|██        | 571/2855 [24:02<1:01:05,  1.61s/it, loss=0.1997, lr=2.84e-05]
Epoch 2/8:  20%|██        | 572/2855 [24:02<1:02:50,  1.65s/it, loss=0.1997, lr=2.84e-05]
Epoch 2/8:  20%|██        | 572/2855 [24:04<1:02:50,  1.65s/it, loss=0.0355, lr=2.84e-05]
Epoch 2/8:  20%|██        | 573/2855 [24:04<1:01:09,  1.61s/it, loss=0.0355, lr=2.84e-05]
Epoch 2/8:  20%|██        | 573/2855 [24:06<1:01:09,  1.61s/it, loss=0.1086, lr=2.84e-05]
Epoch 2/8:  20%|██        | 574/2855 [24:06<1:01:08,  1.61s/it, loss=0.1086, lr=2.84e-05]
Epoch 2/8:  20%|██        | 574/2855 [24:07<1:01:08,  1.61s/it, loss=0.2084, lr=2.84e-05]
Epoch 2/8:  20%|██        | 575/2855 [24:07<1:02:49,  1.65s/it, loss=0.2084, lr=2.84e-05]
Epoch 2/8:  20%|██        | 575/2855 [24:09<1:02:49,  1.65s/it, loss=0.2238, lr=2.84e-05]
Epoch 2/8:  20%|██        | 576/2855 [24:09<1:02:41,  1.65s/it, loss=0.2238, lr=2.84e-05]
Epoch 2/8:  20%|██        | 576/2855 [24:11<1:02:41,  1.65s/it, loss=0.2020, lr=2.84e-05]
Epoch 2/8:  20%|██        | 577/2855 [24:11<1:03:33,  1.67s/it, loss=0.2020, lr=2.84e-05]
Epoch 2/8:  20%|██        | 577/2855 [24:12<1:03:33,  1.67s/it, loss=0.0840, lr=2.84e-05]
Epoch 2/8:  20%|██        | 578/2855 [24:12<1:03:51,  1.68s/it, loss=0.0840, lr=2.84e-05]
Epoch 2/8:  20%|██        | 578/2855 [24:14<1:03:51,  1.68s/it, loss=0.1513, lr=2.84e-05]
Epoch 2/8:  20%|██        | 579/2855 [24:14<1:02:14,  1.64s/it, loss=0.1513, lr=2.84e-05]
Epoch 2/8:  20%|██        | 579/2855 [24:15<1:02:14,  1.64s/it, loss=0.2577, lr=2.84e-05]
Epoch 2/8:  20%|██        | 580/2855 [24:15<1:01:31,  1.62s/it, loss=0.2577, lr=2.84e-05]
Epoch 2/8:  20%|██        | 580/2855 [24:17<1:01:31,  1.62s/it, loss=0.1861, lr=2.84e-05]
Epoch 2/8:  20%|██        | 581/2855 [24:17<1:02:39,  1.65s/it, loss=0.1861, lr=2.84e-05]
Epoch 2/8:  20%|██        | 581/2855 [24:19<1:02:39,  1.65s/it, loss=0.1048, lr=2.84e-05]
Epoch 2/8:  20%|██        | 582/2855 [24:19<1:01:44,  1.63s/it, loss=0.1048, lr=2.84e-05]
Epoch 2/8:  20%|██        | 582/2855 [24:20<1:01:44,  1.63s/it, loss=0.2144, lr=2.84e-05]
Epoch 2/8:  20%|██        | 583/2855 [24:20<1:01:00,  1.61s/it, loss=0.2144, lr=2.84e-05]
Epoch 2/8:  20%|██        | 583/2855 [24:22<1:01:00,  1.61s/it, loss=0.1049, lr=2.84e-05]
Epoch 2/8:  20%|██        | 584/2855 [24:22<1:01:39,  1.63s/it, loss=0.1049, lr=2.84e-05]
Epoch 2/8:  20%|██        | 584/2855 [24:23<1:01:39,  1.63s/it, loss=0.1773, lr=2.84e-05]
Epoch 2/8:  20%|██        | 585/2855 [24:23<58:25,  1.54s/it, loss=0.1773, lr=2.84e-05]  
Epoch 2/8:  20%|██        | 585/2855 [24:25<58:25,  1.54s/it, loss=0.1340, lr=2.84e-05]
Epoch 2/8:  21%|██        | 586/2855 [24:25<56:04,  1.48s/it, loss=0.1340, lr=2.84e-05]
Epoch 2/8:  21%|██        | 586/2855 [24:26<56:04,  1.48s/it, loss=0.1689, lr=2.84e-05]
Epoch 2/8:  21%|██        | 587/2855 [24:26<55:52,  1.48s/it, loss=0.1689, lr=2.84e-05]
Epoch 2/8:  21%|██        | 587/2855 [24:28<55:52,  1.48s/it, loss=0.1122, lr=2.84e-05]
Epoch 2/8:  21%|██        | 588/2855 [24:28<54:46,  1.45s/it, loss=0.1122, lr=2.84e-05]
Epoch 2/8:  21%|██        | 588/2855 [24:29<54:46,  1.45s/it, loss=0.0326, lr=2.84e-05]
Epoch 2/8:  21%|██        | 589/2855 [24:29<54:46,  1.45s/it, loss=0.0326, lr=2.84e-05]
Epoch 2/8:  21%|██        | 589/2855 [24:31<54:46,  1.45s/it, loss=0.2116, lr=2.84e-05]
Epoch 2/8:  21%|██        | 590/2855 [24:31<55:33,  1.47s/it, loss=0.2116, lr=2.84e-05]
Epoch 2/8:  21%|██        | 590/2855 [24:32<55:33,  1.47s/it, loss=0.1545, lr=2.84e-05]
Epoch 2/8:  21%|██        | 591/2855 [24:32<56:11,  1.49s/it, loss=0.1545, lr=2.84e-05]
Epoch 2/8:  21%|██        | 591/2855 [24:33<56:11,  1.49s/it, loss=0.1721, lr=2.84e-05]
Epoch 2/8:  21%|██        | 592/2855 [24:33<54:26,  1.44s/it, loss=0.1721, lr=2.84e-05]
Epoch 2/8:  21%|██        | 592/2855 [24:35<54:26,  1.44s/it, loss=0.1768, lr=2.84e-05]
Epoch 2/8:  21%|██        | 593/2855 [24:35<53:35,  1.42s/it, loss=0.1768, lr=2.84e-05]
Epoch 2/8:  21%|██        | 593/2855 [24:36<53:35,  1.42s/it, loss=0.2618, lr=2.84e-05]
Epoch 2/8:  21%|██        | 594/2855 [24:36<56:18,  1.49s/it, loss=0.2618, lr=2.84e-05]
Epoch 2/8:  21%|██        | 594/2855 [24:38<56:18,  1.49s/it, loss=0.1894, lr=2.84e-05]
Epoch 2/8:  21%|██        | 595/2855 [24:38<59:07,  1.57s/it, loss=0.1894, lr=2.84e-05]
Epoch 2/8:  21%|██        | 595/2855 [24:40<59:07,  1.57s/it, loss=0.0750, lr=2.84e-05]
Epoch 2/8:  21%|██        | 596/2855 [24:40<1:00:26,  1.61s/it, loss=0.0750, lr=2.84e-05]
Epoch 2/8:  21%|██        | 596/2855 [24:42<1:00:26,  1.61s/it, loss=0.1078, lr=2.84e-05]
Epoch 2/8:  21%|██        | 597/2855 [24:42<1:01:39,  1.64s/it, loss=0.1078, lr=2.84e-05]
Epoch 2/8:  21%|██        | 597/2855 [24:43<1:01:39,  1.64s/it, loss=0.2033, lr=2.84e-05]
Epoch 2/8:  21%|██        | 598/2855 [24:43<1:01:17,  1.63s/it, loss=0.2033, lr=2.84e-05]
Epoch 2/8:  21%|██        | 598/2855 [24:45<1:01:17,  1.63s/it, loss=0.0691, lr=2.84e-05]
Epoch 2/8:  21%|██        | 599/2855 [24:45<1:00:21,  1.61s/it, loss=0.0691, lr=2.84e-05]
Epoch 2/8:  21%|██        | 599/2855 [24:46<1:00:21,  1.61s/it, loss=0.2394, lr=2.84e-05]
Epoch 2/8:  21%|██        | 600/2855 [24:46<1:00:41,  1.61s/it, loss=0.2394, lr=2.84e-05]
Epoch 2/8:  21%|██        | 600/2855 [24:48<1:00:41,  1.61s/it, loss=0.1894, lr=2.84e-05]
Epoch 2/8:  21%|██        | 601/2855 [24:48<1:01:57,  1.65s/it, loss=0.1894, lr=2.84e-05]
Epoch 2/8:  21%|██        | 601/2855 [24:50<1:01:57,  1.65s/it, loss=0.0926, lr=2.84e-05]
Epoch 2/8:  21%|██        | 602/2855 [24:50<1:02:10,  1.66s/it, loss=0.0926, lr=2.84e-05]
Epoch 2/8:  21%|██        | 602/2855 [24:52<1:02:10,  1.66s/it, loss=0.1929, lr=2.84e-05]
Epoch 2/8:  21%|██        | 603/2855 [24:52<1:02:47,  1.67s/it, loss=0.1929, lr=2.84e-05]
Epoch 2/8:  21%|██        | 603/2855 [24:53<1:02:47,  1.67s/it, loss=0.1857, lr=2.84e-05]
Epoch 2/8:  21%|██        | 604/2855 [24:53<1:02:21,  1.66s/it, loss=0.1857, lr=2.84e-05]
Epoch 2/8:  21%|██        | 604/2855 [24:55<1:02:21,  1.66s/it, loss=0.1618, lr=2.84e-05]
Epoch 2/8:  21%|██        | 605/2855 [24:55<1:00:50,  1.62s/it, loss=0.1618, lr=2.84e-05]
Epoch 2/8:  21%|██        | 605/2855 [24:56<1:00:50,  1.62s/it, loss=0.1062, lr=2.84e-05]
Epoch 2/8:  21%|██        | 606/2855 [24:56<1:00:51,  1.62s/it, loss=0.1062, lr=2.84e-05]
Epoch 2/8:  21%|██        | 606/2855 [24:58<1:00:51,  1.62s/it, loss=0.2706, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 607/2855 [24:58<59:10,  1.58s/it, loss=0.2706, lr=2.84e-05]  
Epoch 2/8:  21%|██▏       | 607/2855 [24:59<59:10,  1.58s/it, loss=0.2045, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 608/2855 [24:59<58:21,  1.56s/it, loss=0.2045, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 608/2855 [25:01<58:21,  1.56s/it, loss=0.1909, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 609/2855 [25:01<58:13,  1.56s/it, loss=0.1909, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 609/2855 [25:02<58:13,  1.56s/it, loss=0.1023, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 610/2855 [25:02<57:55,  1.55s/it, loss=0.1023, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 610/2855 [25:04<57:55,  1.55s/it, loss=0.1270, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 611/2855 [25:04<55:54,  1.50s/it, loss=0.1270, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 611/2855 [25:05<55:54,  1.50s/it, loss=0.1592, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 612/2855 [25:05<55:20,  1.48s/it, loss=0.1592, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 612/2855 [25:07<55:20,  1.48s/it, loss=0.1798, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 613/2855 [25:07<55:59,  1.50s/it, loss=0.1798, lr=2.84e-05]
Epoch 2/8:  21%|██▏       | 613/2855 [25:08<55:59,  1.50s/it, loss=0.2053, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 614/2855 [25:08<54:57,  1.47s/it, loss=0.2053, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 614/2855 [25:10<54:57,  1.47s/it, loss=0.1096, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 615/2855 [25:10<54:03,  1.45s/it, loss=0.1096, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 615/2855 [25:11<54:03,  1.45s/it, loss=0.0752, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 616/2855 [25:11<57:24,  1.54s/it, loss=0.0752, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 616/2855 [25:13<57:24,  1.54s/it, loss=0.2119, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 617/2855 [25:13<1:00:26,  1.62s/it, loss=0.2119, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 617/2855 [25:15<1:00:26,  1.62s/it, loss=0.1614, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 618/2855 [25:15<1:00:29,  1.62s/it, loss=0.1614, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 618/2855 [25:16<1:00:29,  1.62s/it, loss=0.0866, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 619/2855 [25:16<1:00:23,  1.62s/it, loss=0.0866, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 619/2855 [25:18<1:00:23,  1.62s/it, loss=0.1386, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 620/2855 [25:18<1:01:01,  1.64s/it, loss=0.1386, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 620/2855 [25:20<1:01:01,  1.64s/it, loss=0.2781, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 621/2855 [25:20<1:00:58,  1.64s/it, loss=0.2781, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 621/2855 [25:21<1:00:58,  1.64s/it, loss=0.0637, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 622/2855 [25:21<1:01:34,  1.65s/it, loss=0.0637, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 622/2855 [25:23<1:01:34,  1.65s/it, loss=0.1111, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 623/2855 [25:23<1:02:48,  1.69s/it, loss=0.1111, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 623/2855 [25:25<1:02:48,  1.69s/it, loss=0.2094, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 624/2855 [25:25<1:01:27,  1.65s/it, loss=0.2094, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 624/2855 [25:26<1:01:27,  1.65s/it, loss=0.0392, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 625/2855 [25:26<1:01:12,  1.65s/it, loss=0.0392, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 625/2855 [25:28<1:01:12,  1.65s/it, loss=0.1159, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 626/2855 [25:28<1:01:27,  1.65s/it, loss=0.1159, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 626/2855 [25:30<1:01:27,  1.65s/it, loss=0.1226, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 627/2855 [25:30<1:01:59,  1.67s/it, loss=0.1226, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 627/2855 [25:31<1:01:59,  1.67s/it, loss=0.1165, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 628/2855 [25:31<1:02:27,  1.68s/it, loss=0.1165, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 628/2855 [25:33<1:02:27,  1.68s/it, loss=0.1465, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 629/2855 [25:33<1:01:17,  1.65s/it, loss=0.1465, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 629/2855 [25:34<1:01:17,  1.65s/it, loss=0.1825, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 630/2855 [25:34<57:31,  1.55s/it, loss=0.1825, lr=2.84e-05]  
Epoch 2/8:  22%|██▏       | 630/2855 [25:36<57:31,  1.55s/it, loss=0.1392, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 631/2855 [25:36<54:58,  1.48s/it, loss=0.1392, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 631/2855 [25:37<54:58,  1.48s/it, loss=0.0799, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 632/2855 [25:37<55:22,  1.49s/it, loss=0.0799, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 632/2855 [25:39<55:22,  1.49s/it, loss=0.2814, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 633/2855 [25:39<58:29,  1.58s/it, loss=0.2814, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 633/2855 [25:41<58:29,  1.58s/it, loss=0.1057, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 634/2855 [25:41<59:50,  1.62s/it, loss=0.1057, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 634/2855 [25:42<59:50,  1.62s/it, loss=0.2771, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 635/2855 [25:42<1:00:34,  1.64s/it, loss=0.2771, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 635/2855 [25:44<1:00:34,  1.64s/it, loss=0.1002, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 636/2855 [25:44<59:58,  1.62s/it, loss=0.1002, lr=2.84e-05]  
Epoch 2/8:  22%|██▏       | 636/2855 [25:45<59:58,  1.62s/it, loss=0.0553, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 637/2855 [25:45<58:58,  1.60s/it, loss=0.0553, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 637/2855 [25:47<58:58,  1.60s/it, loss=0.1799, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 638/2855 [25:47<59:53,  1.62s/it, loss=0.1799, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 638/2855 [25:49<59:53,  1.62s/it, loss=0.2940, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 639/2855 [25:49<1:01:35,  1.67s/it, loss=0.2940, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 639/2855 [25:51<1:01:35,  1.67s/it, loss=0.1003, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 640/2855 [25:51<1:01:07,  1.66s/it, loss=0.1003, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 640/2855 [25:52<1:01:07,  1.66s/it, loss=0.0702, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 641/2855 [25:52<1:02:00,  1.68s/it, loss=0.0702, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 641/2855 [25:54<1:02:00,  1.68s/it, loss=0.1154, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 642/2855 [25:54<1:01:37,  1.67s/it, loss=0.1154, lr=2.84e-05]
Epoch 2/8:  22%|██▏       | 642/2855 [25:55<1:01:37,  1.67s/it, loss=0.1653, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 643/2855 [25:55<59:48,  1.62s/it, loss=0.1653, lr=2.84e-05]  
Epoch 2/8:  23%|██▎       | 643/2855 [25:57<59:48,  1.62s/it, loss=0.3195, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 644/2855 [25:57<59:32,  1.62s/it, loss=0.3195, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 644/2855 [25:58<59:32,  1.62s/it, loss=0.0971, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 645/2855 [25:58<57:58,  1.57s/it, loss=0.0971, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 645/2855 [26:00<57:58,  1.57s/it, loss=0.1827, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 646/2855 [26:00<56:16,  1.53s/it, loss=0.1827, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 646/2855 [26:01<56:16,  1.53s/it, loss=0.2199, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 647/2855 [26:01<55:38,  1.51s/it, loss=0.2199, lr=2.84e-05]
Epoch 2/8:  23%|██▎       | 647/2855 [26:03<55:38,  1.51s/it, loss=0.1427, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 648/2855 [26:03<57:20,  1.56s/it, loss=0.1427, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 648/2855 [26:05<57:20,  1.56s/it, loss=0.2101, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 649/2855 [26:05<56:40,  1.54s/it, loss=0.2101, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 649/2855 [26:06<56:40,  1.54s/it, loss=0.1593, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 650/2855 [26:06<54:54,  1.49s/it, loss=0.1593, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 650/2855 [26:07<54:54,  1.49s/it, loss=0.1517, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 651/2855 [26:07<54:41,  1.49s/it, loss=0.1517, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 651/2855 [26:09<54:41,  1.49s/it, loss=0.1759, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 652/2855 [26:09<53:34,  1.46s/it, loss=0.1759, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 652/2855 [26:10<53:34,  1.46s/it, loss=0.0697, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 653/2855 [26:10<54:55,  1.50s/it, loss=0.0697, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 653/2855 [26:12<54:55,  1.50s/it, loss=0.2348, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 654/2855 [26:12<59:37,  1.63s/it, loss=0.2348, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 654/2855 [26:14<59:37,  1.63s/it, loss=0.1321, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 655/2855 [26:14<1:00:26,  1.65s/it, loss=0.1321, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 655/2855 [26:16<1:00:26,  1.65s/it, loss=0.2321, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 656/2855 [26:16<58:47,  1.60s/it, loss=0.2321, lr=2.83e-05]  
Epoch 2/8:  23%|██▎       | 656/2855 [26:17<58:47,  1.60s/it, loss=0.0376, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 657/2855 [26:17<58:53,  1.61s/it, loss=0.0376, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 657/2855 [26:19<58:53,  1.61s/it, loss=0.1009, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 658/2855 [26:19<1:00:18,  1.65s/it, loss=0.1009, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 658/2855 [26:21<1:00:18,  1.65s/it, loss=0.1231, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 659/2855 [26:21<1:00:25,  1.65s/it, loss=0.1231, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 659/2855 [26:22<1:00:25,  1.65s/it, loss=0.1143, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 660/2855 [26:22<1:01:29,  1.68s/it, loss=0.1143, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 660/2855 [26:24<1:01:29,  1.68s/it, loss=0.0641, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 661/2855 [26:24<1:01:30,  1.68s/it, loss=0.0641, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 661/2855 [26:26<1:01:30,  1.68s/it, loss=0.1872, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 662/2855 [26:26<1:00:04,  1.64s/it, loss=0.1872, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 662/2855 [26:27<1:00:04,  1.64s/it, loss=0.1628, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 663/2855 [26:27<59:47,  1.64s/it, loss=0.1628, lr=2.83e-05]  
Epoch 2/8:  23%|██▎       | 663/2855 [26:29<59:47,  1.64s/it, loss=0.1716, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 664/2855 [26:29<59:55,  1.64s/it, loss=0.1716, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 664/2855 [26:30<59:55,  1.64s/it, loss=0.1762, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 665/2855 [26:30<59:59,  1.64s/it, loss=0.1762, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 665/2855 [26:32<59:59,  1.64s/it, loss=0.1021, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 666/2855 [26:32<59:25,  1.63s/it, loss=0.1021, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 666/2855 [26:34<59:25,  1.63s/it, loss=0.2100, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 667/2855 [26:34<59:36,  1.63s/it, loss=0.2100, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 667/2855 [26:35<59:36,  1.63s/it, loss=0.1552, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 668/2855 [26:35<56:45,  1.56s/it, loss=0.1552, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 668/2855 [26:36<56:45,  1.56s/it, loss=0.0533, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 669/2855 [26:36<55:08,  1.51s/it, loss=0.0533, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 669/2855 [26:38<55:08,  1.51s/it, loss=0.2366, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 670/2855 [26:38<55:00,  1.51s/it, loss=0.2366, lr=2.83e-05]
Epoch 2/8:  23%|██▎       | 670/2855 [26:39<55:00,  1.51s/it, loss=0.1094, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 671/2855 [26:39<53:37,  1.47s/it, loss=0.1094, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 671/2855 [26:41<53:37,  1.47s/it, loss=0.1438, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 672/2855 [26:41<52:29,  1.44s/it, loss=0.1438, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 672/2855 [26:42<52:29,  1.44s/it, loss=0.1620, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 673/2855 [26:42<53:50,  1.48s/it, loss=0.1620, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 673/2855 [26:44<53:50,  1.48s/it, loss=0.1474, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 674/2855 [26:44<55:02,  1.51s/it, loss=0.1474, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 674/2855 [26:46<55:02,  1.51s/it, loss=0.2548, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 675/2855 [26:46<56:26,  1.55s/it, loss=0.2548, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 675/2855 [26:47<56:26,  1.55s/it, loss=0.2269, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 676/2855 [26:47<58:10,  1.60s/it, loss=0.2269, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 676/2855 [26:49<58:10,  1.60s/it, loss=0.1035, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 677/2855 [26:49<1:00:04,  1.66s/it, loss=0.1035, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 677/2855 [26:51<1:00:04,  1.66s/it, loss=0.1769, lr=2.83e-05]
Epoch 2/8:  24%|██▎       | 678/2855 [26:51<59:16,  1.63s/it, loss=0.1769, lr=2.83e-05]  
Epoch 2/8:  24%|██▎       | 678/2855 [26:52<59:16,  1.63s/it, loss=0.1468, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 679/2855 [26:52<59:13,  1.63s/it, loss=0.1468, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 679/2855 [26:54<59:13,  1.63s/it, loss=0.2460, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 680/2855 [26:54<59:47,  1.65s/it, loss=0.2460, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 680/2855 [26:55<59:47,  1.65s/it, loss=0.1577, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 681/2855 [26:55<58:31,  1.62s/it, loss=0.1577, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 681/2855 [26:57<58:31,  1.62s/it, loss=0.2224, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 682/2855 [26:57<58:25,  1.61s/it, loss=0.2224, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 682/2855 [26:59<58:25,  1.61s/it, loss=0.1642, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 683/2855 [26:59<58:49,  1.62s/it, loss=0.1642, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 683/2855 [27:00<58:49,  1.62s/it, loss=0.1479, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 684/2855 [27:00<59:07,  1.63s/it, loss=0.1479, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 684/2855 [27:02<59:07,  1.63s/it, loss=0.1939, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 685/2855 [27:02<1:00:01,  1.66s/it, loss=0.1939, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 685/2855 [27:04<1:00:01,  1.66s/it, loss=0.2441, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 686/2855 [27:04<1:01:29,  1.70s/it, loss=0.2441, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 686/2855 [27:05<1:01:29,  1.70s/it, loss=0.1076, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 687/2855 [27:05<58:12,  1.61s/it, loss=0.1076, lr=2.83e-05]  
Epoch 2/8:  24%|██▍       | 687/2855 [27:07<58:12,  1.61s/it, loss=0.3268, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 688/2855 [27:07<55:11,  1.53s/it, loss=0.3268, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 688/2855 [27:08<55:11,  1.53s/it, loss=0.0827, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 689/2855 [27:08<54:23,  1.51s/it, loss=0.0827, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 689/2855 [27:09<54:23,  1.51s/it, loss=0.0830, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 690/2855 [27:09<52:26,  1.45s/it, loss=0.0830, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 690/2855 [27:11<52:26,  1.45s/it, loss=0.1187, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 691/2855 [27:11<51:44,  1.43s/it, loss=0.1187, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 691/2855 [27:13<51:44,  1.43s/it, loss=0.0607, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 692/2855 [27:13<55:16,  1.53s/it, loss=0.0607, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 692/2855 [27:14<55:16,  1.53s/it, loss=0.0012, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 693/2855 [27:14<58:10,  1.61s/it, loss=0.0012, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 693/2855 [27:16<58:10,  1.61s/it, loss=0.0604, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 694/2855 [27:16<57:07,  1.59s/it, loss=0.0604, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 694/2855 [27:17<57:07,  1.59s/it, loss=0.0011, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 695/2855 [27:17<56:49,  1.58s/it, loss=0.0011, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 695/2855 [27:19<56:49,  1.58s/it, loss=0.1041, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 696/2855 [27:19<58:02,  1.61s/it, loss=0.1041, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 696/2855 [27:21<58:02,  1.61s/it, loss=0.0882, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 697/2855 [27:21<58:06,  1.62s/it, loss=0.0882, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 697/2855 [27:22<58:06,  1.62s/it, loss=0.1581, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 698/2855 [27:22<58:50,  1.64s/it, loss=0.1581, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 698/2855 [27:24<58:50,  1.64s/it, loss=0.2032, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 699/2855 [27:24<59:31,  1.66s/it, loss=0.2032, lr=2.83e-05]
Epoch 2/8:  24%|██▍       | 699/2855 [27:26<59:31,  1.66s/it, loss=0.0786, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 700/2855 [27:26<58:48,  1.64s/it, loss=0.0786, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 700/2855 [27:27<58:48,  1.64s/it, loss=0.0913, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 701/2855 [27:27<58:12,  1.62s/it, loss=0.0913, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 701/2855 [27:29<58:12,  1.62s/it, loss=0.2073, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 702/2855 [27:29<59:23,  1.66s/it, loss=0.2073, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 702/2855 [27:31<59:23,  1.66s/it, loss=0.1966, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 703/2855 [27:31<59:28,  1.66s/it, loss=0.1966, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 703/2855 [27:32<59:28,  1.66s/it, loss=0.0984, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 704/2855 [27:32<58:25,  1.63s/it, loss=0.0984, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 704/2855 [27:34<58:25,  1.63s/it, loss=0.1181, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 705/2855 [27:34<58:38,  1.64s/it, loss=0.1181, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 705/2855 [27:35<58:38,  1.64s/it, loss=0.2120, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 706/2855 [27:35<55:09,  1.54s/it, loss=0.2120, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 706/2855 [27:37<55:09,  1.54s/it, loss=0.1690, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 707/2855 [27:37<53:04,  1.48s/it, loss=0.1690, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 707/2855 [27:38<53:04,  1.48s/it, loss=0.1426, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 708/2855 [27:38<53:15,  1.49s/it, loss=0.1426, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 708/2855 [27:40<53:15,  1.49s/it, loss=0.1295, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 709/2855 [27:40<52:30,  1.47s/it, loss=0.1295, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 709/2855 [27:41<52:30,  1.47s/it, loss=0.1359, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 710/2855 [27:41<51:10,  1.43s/it, loss=0.1359, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 710/2855 [27:42<51:10,  1.43s/it, loss=0.0392, lr=2.83e-05]
Epoch 2/8:  25%|���█▍       | 711/2855 [27:42<51:06,  1.43s/it, loss=0.0392, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 711/2855 [27:44<51:06,  1.43s/it, loss=0.0004, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 712/2855 [27:44<52:58,  1.48s/it, loss=0.0004, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 712/2855 [27:46<52:58,  1.48s/it, loss=0.0837, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 713/2855 [27:46<54:48,  1.54s/it, loss=0.0837, lr=2.83e-05]
Epoch 2/8:  25%|██▍       | 713/2855 [27:47<54:48,  1.54s/it, loss=0.1294, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 714/2855 [27:47<56:09,  1.57s/it, loss=0.1294, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 714/2855 [27:49<56:09,  1.57s/it, loss=0.1311, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 715/2855 [27:49<58:29,  1.64s/it, loss=0.1311, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 715/2855 [27:51<58:29,  1.64s/it, loss=0.1837, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 716/2855 [27:51<58:36,  1.64s/it, loss=0.1837, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 716/2855 [27:52<58:36,  1.64s/it, loss=0.1032, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 717/2855 [27:52<58:06,  1.63s/it, loss=0.1032, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 717/2855 [27:54<58:06,  1.63s/it, loss=0.1694, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 718/2855 [27:54<59:15,  1.66s/it, loss=0.1694, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 718/2855 [27:56<59:15,  1.66s/it, loss=0.0940, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 719/2855 [27:56<58:31,  1.64s/it, loss=0.0940, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 719/2855 [27:57<58:31,  1.64s/it, loss=0.1282, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 720/2855 [27:57<56:54,  1.60s/it, loss=0.1282, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 720/2855 [27:59<56:54,  1.60s/it, loss=0.2572, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 721/2855 [27:59<57:23,  1.61s/it, loss=0.2572, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 721/2855 [28:00<57:23,  1.61s/it, loss=0.0783, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 722/2855 [28:00<58:13,  1.64s/it, loss=0.0783, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 722/2855 [28:02<58:13,  1.64s/it, loss=0.1821, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 723/2855 [28:02<58:38,  1.65s/it, loss=0.1821, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 723/2855 [28:04<58:38,  1.65s/it, loss=0.1617, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 724/2855 [28:04<59:27,  1.67s/it, loss=0.1617, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 724/2855 [28:05<59:27,  1.67s/it, loss=0.1486, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 725/2855 [28:05<56:51,  1.60s/it, loss=0.1486, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 725/2855 [28:07<56:51,  1.60s/it, loss=0.0811, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 726/2855 [28:07<53:58,  1.52s/it, loss=0.0811, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 726/2855 [28:08<53:58,  1.52s/it, loss=0.1482, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 727/2855 [28:08<53:31,  1.51s/it, loss=0.1482, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 727/2855 [28:10<53:31,  1.51s/it, loss=0.1777, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 728/2855 [28:10<53:00,  1.50s/it, loss=0.1777, lr=2.83e-05]
Epoch 2/8:  25%|██▌       | 728/2855 [28:11<53:00,  1.50s/it, loss=0.1156, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 729/2855 [28:11<51:16,  1.45s/it, loss=0.1156, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 729/2855 [28:12<51:16,  1.45s/it, loss=0.2357, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 730/2855 [28:12<50:48,  1.43s/it, loss=0.2357, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 730/2855 [28:14<50:48,  1.43s/it, loss=0.1029, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 731/2855 [28:14<52:47,  1.49s/it, loss=0.1029, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 731/2855 [28:15<52:47,  1.49s/it, loss=0.2040, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 732/2855 [28:15<51:23,  1.45s/it, loss=0.2040, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 732/2855 [28:17<51:23,  1.45s/it, loss=0.1896, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 733/2855 [28:17<49:44,  1.41s/it, loss=0.1896, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 733/2855 [28:18<49:44,  1.41s/it, loss=0.3013, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 734/2855 [28:18<50:35,  1.43s/it, loss=0.3013, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 734/2855 [28:20<50:35,  1.43s/it, loss=0.1748, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 735/2855 [28:20<53:25,  1.51s/it, loss=0.1748, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 735/2855 [28:22<53:25,  1.51s/it, loss=0.0776, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 736/2855 [28:22<55:45,  1.58s/it, loss=0.0776, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 736/2855 [28:23<55:45,  1.58s/it, loss=0.1709, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 737/2855 [28:23<58:06,  1.65s/it, loss=0.1709, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 737/2855 [28:25<58:06,  1.65s/it, loss=0.1806, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 738/2855 [28:25<58:25,  1.66s/it, loss=0.1806, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 738/2855 [28:27<58:25,  1.66s/it, loss=0.2262, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 739/2855 [28:27<56:57,  1.61s/it, loss=0.2262, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 739/2855 [28:28<56:57,  1.61s/it, loss=0.1397, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 740/2855 [28:28<56:47,  1.61s/it, loss=0.1397, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 740/2855 [28:30<56:47,  1.61s/it, loss=0.1648, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 741/2855 [28:30<57:56,  1.64s/it, loss=0.1648, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 741/2855 [28:32<57:56,  1.64s/it, loss=0.2439, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 742/2855 [28:32<59:31,  1.69s/it, loss=0.2439, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 742/2855 [28:33<59:31,  1.69s/it, loss=0.2361, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 743/2855 [28:33<1:00:01,  1.71s/it, loss=0.2361, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 743/2855 [28:35<1:00:01,  1.71s/it, loss=0.1496, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 744/2855 [28:35<1:02:46,  1.78s/it, loss=0.1496, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 744/2855 [28:37<1:02:46,  1.78s/it, loss=0.1360, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 745/2855 [28:37<1:04:57,  1.85s/it, loss=0.1360, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 745/2855 [28:39<1:04:57,  1.85s/it, loss=0.2157, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 746/2855 [28:39<1:04:26,  1.83s/it, loss=0.2157, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 746/2855 [28:41<1:04:26,  1.83s/it, loss=0.1302, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 747/2855 [28:41<1:07:58,  1.93s/it, loss=0.1302, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 747/2855 [28:43<1:07:58,  1.93s/it, loss=0.1395, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 748/2855 [28:43<1:07:18,  1.92s/it, loss=0.1395, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 748/2855 [28:45<1:07:18,  1.92s/it, loss=0.1381, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 749/2855 [28:45<1:05:46,  1.87s/it, loss=0.1381, lr=2.83e-05]
Epoch 2/8:  26%|██▌       | 749/2855 [28:46<1:05:46,  1.87s/it, loss=0.1123, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 750/2855 [28:46<1:01:31,  1.75s/it, loss=0.1123, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 750/2855 [28:48<1:01:31,  1.75s/it, loss=0.1391, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 751/2855 [28:48<58:43,  1.67s/it, loss=0.1391, lr=2.83e-05]  
Epoch 2/8:  26%|██▋       | 751/2855 [28:50<58:43,  1.67s/it, loss=0.1326, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 752/2855 [28:50<1:02:00,  1.77s/it, loss=0.1326, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 752/2855 [28:52<1:02:00,  1.77s/it, loss=0.1275, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 753/2855 [28:52<1:06:25,  1.90s/it, loss=0.1275, lr=2.83e-05]
Epoch 2/8:  26%|██▋       | 753/2855 [28:54<1:06:25,  1.90s/it, loss=0.1646, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 754/2855 [28:54<1:05:29,  1.87s/it, loss=0.1646, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 754/2855 [28:56<1:05:29,  1.87s/it, loss=0.1216, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 755/2855 [28:56<1:07:10,  1.92s/it, loss=0.1216, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 755/2855 [28:58<1:07:10,  1.92s/it, loss=0.1751, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 756/2855 [28:58<1:04:02,  1.83s/it, loss=0.1751, lr=2.82e-05]
Epoch 2/8:  26%|██▋       | 756/2855 [28:59<1:04:02,  1.83s/it, loss=0.0468, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 757/2855 [28:59<1:03:05,  1.80s/it, loss=0.0468, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 757/2855 [29:01<1:03:05,  1.80s/it, loss=0.2451, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 758/2855 [29:01<1:05:14,  1.87s/it, loss=0.2451, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 758/2855 [29:03<1:05:14,  1.87s/it, loss=0.1058, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 759/2855 [29:03<1:04:28,  1.85s/it, loss=0.1058, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 759/2855 [29:05<1:04:28,  1.85s/it, loss=0.1743, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 760/2855 [29:05<1:05:33,  1.88s/it, loss=0.1743, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 760/2855 [29:07<1:05:33,  1.88s/it, loss=0.1549, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 761/2855 [29:07<1:07:27,  1.93s/it, loss=0.1549, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 761/2855 [29:09<1:07:27,  1.93s/it, loss=0.1960, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 762/2855 [29:09<1:08:25,  1.96s/it, loss=0.1960, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 762/2855 [29:11<1:08:25,  1.96s/it, loss=0.2010, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 763/2855 [29:11<1:10:30,  2.02s/it, loss=0.2010, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 763/2855 [29:13<1:10:30,  2.02s/it, loss=0.2188, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 764/2855 [29:13<1:08:12,  1.96s/it, loss=0.2188, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 764/2855 [29:15<1:08:12,  1.96s/it, loss=0.1450, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 765/2855 [29:15<1:06:49,  1.92s/it, loss=0.1450, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 765/2855 [29:18<1:06:49,  1.92s/it, loss=0.2791, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 766/2855 [29:18<1:17:22,  2.22s/it, loss=0.2791, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 766/2855 [29:21<1:17:22,  2.22s/it, loss=0.1691, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 767/2855 [29:21<1:25:05,  2.45s/it, loss=0.1691, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 767/2855 [29:23<1:25:05,  2.45s/it, loss=0.2100, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 768/2855 [29:23<1:21:35,  2.35s/it, loss=0.2100, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 768/2855 [29:25<1:21:35,  2.35s/it, loss=0.1210, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 769/2855 [29:25<1:17:30,  2.23s/it, loss=0.1210, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 769/2855 [29:27<1:17:30,  2.23s/it, loss=0.1640, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 770/2855 [29:27<1:17:59,  2.24s/it, loss=0.1640, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 770/2855 [29:29<1:17:59,  2.24s/it, loss=0.2436, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 771/2855 [29:29<1:15:56,  2.19s/it, loss=0.2436, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 771/2855 [29:32<1:15:56,  2.19s/it, loss=0.1835, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 772/2855 [29:32<1:17:12,  2.22s/it, loss=0.1835, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 772/2855 [29:34<1:17:12,  2.22s/it, loss=0.1349, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 773/2855 [29:34<1:18:13,  2.25s/it, loss=0.1349, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 773/2855 [29:36<1:18:13,  2.25s/it, loss=0.0620, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 774/2855 [29:36<1:19:17,  2.29s/it, loss=0.0620, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 774/2855 [29:39<1:19:17,  2.29s/it, loss=0.0774, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 775/2855 [29:39<1:21:54,  2.36s/it, loss=0.0774, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 775/2855 [29:41<1:21:54,  2.36s/it, loss=0.0630, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 776/2855 [29:41<1:22:07,  2.37s/it, loss=0.0630, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 776/2855 [29:44<1:22:07,  2.37s/it, loss=0.1702, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 777/2855 [29:44<1:21:42,  2.36s/it, loss=0.1702, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 777/2855 [29:46<1:21:42,  2.36s/it, loss=0.0919, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 778/2855 [29:46<1:22:32,  2.38s/it, loss=0.0919, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 778/2855 [29:48<1:22:32,  2.38s/it, loss=0.2043, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 779/2855 [29:48<1:18:46,  2.28s/it, loss=0.2043, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 779/2855 [29:50<1:18:46,  2.28s/it, loss=0.1683, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 780/2855 [29:50<1:16:19,  2.21s/it, loss=0.1683, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 780/2855 [29:57<1:16:19,  2.21s/it, loss=0.1216, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 781/2855 [29:57<2:07:04,  3.68s/it, loss=0.1216, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 781/2855 [29:59<2:07:04,  3.68s/it, loss=0.0955, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 782/2855 [29:59<1:45:15,  3.05s/it, loss=0.0955, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 782/2855 [30:02<1:45:15,  3.05s/it, loss=0.1214, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 783/2855 [30:02<1:45:12,  3.05s/it, loss=0.1214, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 783/2855 [30:05<1:45:12,  3.05s/it, loss=0.1801, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 784/2855 [30:05<1:41:55,  2.95s/it, loss=0.1801, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 784/2855 [30:07<1:41:55,  2.95s/it, loss=0.0762, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 785/2855 [30:07<1:34:18,  2.73s/it, loss=0.0762, lr=2.82e-05]
Epoch 2/8:  27%|██▋       | 785/2855 [30:09<1:34:18,  2.73s/it, loss=0.1392, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 786/2855 [30:09<1:28:06,  2.56s/it, loss=0.1392, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 786/2855 [30:11<1:28:06,  2.56s/it, loss=0.1274, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 787/2855 [30:11<1:23:12,  2.41s/it, loss=0.1274, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 787/2855 [30:13<1:23:12,  2.41s/it, loss=0.0923, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 788/2855 [30:13<1:23:54,  2.44s/it, loss=0.0923, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 788/2855 [30:16<1:23:54,  2.44s/it, loss=0.3424, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 789/2855 [30:16<1:24:53,  2.47s/it, loss=0.3424, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 789/2855 [30:19<1:24:53,  2.47s/it, loss=0.1213, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 790/2855 [30:19<1:26:36,  2.52s/it, loss=0.1213, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 790/2855 [30:21<1:26:36,  2.52s/it, loss=0.1229, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 791/2855 [30:21<1:21:14,  2.36s/it, loss=0.1229, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 791/2855 [30:23<1:21:14,  2.36s/it, loss=0.1959, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 792/2855 [30:23<1:24:11,  2.45s/it, loss=0.1959, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 792/2855 [30:26<1:24:11,  2.45s/it, loss=0.2461, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 793/2855 [30:26<1:25:18,  2.48s/it, loss=0.2461, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 793/2855 [30:28<1:25:18,  2.48s/it, loss=0.2324, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 794/2855 [30:28<1:18:29,  2.29s/it, loss=0.2324, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 794/2855 [30:30<1:18:29,  2.29s/it, loss=0.2381, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 795/2855 [30:30<1:21:38,  2.38s/it, loss=0.2381, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 795/2855 [31:32<1:21:38,  2.38s/it, loss=0.0995, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 796/2855 [31:32<11:36:04, 20.28s/it, loss=0.0995, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 796/2855 [31:34<11:36:04, 20.28s/it, loss=0.0726, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 797/2855 [31:34<8:22:09, 14.64s/it, loss=0.0726, lr=2.82e-05] 
Epoch 2/8:  28%|██▊       | 797/2855 [31:36<8:22:09, 14.64s/it, loss=0.2328, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 798/2855 [31:36<6:09:16, 10.77s/it, loss=0.2328, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 798/2855 [31:37<6:09:16, 10.77s/it, loss=0.2143, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 799/2855 [31:37<4:36:20,  8.06s/it, loss=0.2143, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 799/2855 [31:39<4:36:20,  8.06s/it, loss=0.0296, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 800/2855 [31:39<3:31:13,  6.17s/it, loss=0.0296, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 800/2855 [31:40<3:31:13,  6.17s/it, loss=0.1566, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 801/2855 [31:40<2:42:32,  4.75s/it, loss=0.1566, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 801/2855 [31:42<2:42:32,  4.75s/it, loss=0.0278, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 802/2855 [31:42<2:09:10,  3.78s/it, loss=0.0278, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 802/2855 [31:44<2:09:10,  3.78s/it, loss=0.2204, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 803/2855 [31:44<1:48:08,  3.16s/it, loss=0.2204, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 803/2855 [31:45<1:48:08,  3.16s/it, loss=0.1748, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 804/2855 [31:45<1:29:42,  2.62s/it, loss=0.1748, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 804/2855 [31:46<1:29:42,  2.62s/it, loss=0.0997, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 805/2855 [31:46<1:16:32,  2.24s/it, loss=0.0997, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 805/2855 [31:48<1:16:32,  2.24s/it, loss=0.1261, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 806/2855 [31:48<1:11:03,  2.08s/it, loss=0.1261, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 806/2855 [31:50<1:11:03,  2.08s/it, loss=0.0625, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 807/2855 [31:50<1:05:01,  1.91s/it, loss=0.0625, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 807/2855 [31:51<1:05:01,  1.91s/it, loss=0.2238, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 808/2855 [31:51<1:03:55,  1.87s/it, loss=0.2238, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 808/2855 [31:54<1:03:55,  1.87s/it, loss=0.1453, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 809/2855 [31:54<1:06:28,  1.95s/it, loss=0.1453, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 809/2855 [31:56<1:06:28,  1.95s/it, loss=0.0626, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 810/2855 [31:56<1:06:40,  1.96s/it, loss=0.0626, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 810/2855 [31:57<1:06:40,  1.96s/it, loss=0.0702, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 811/2855 [31:57<1:05:18,  1.92s/it, loss=0.0702, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 811/2855 [31:59<1:05:18,  1.92s/it, loss=0.1729, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 812/2855 [31:59<1:05:41,  1.93s/it, loss=0.1729, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 812/2855 [32:01<1:05:41,  1.93s/it, loss=0.0578, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 813/2855 [32:01<1:02:14,  1.83s/it, loss=0.0578, lr=2.82e-05]
Epoch 2/8:  28%|██▊       | 813/2855 [32:02<1:02:14,  1.83s/it, loss=0.1197, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 814/2855 [32:02<58:58,  1.73s/it, loss=0.1197, lr=2.82e-05]  
Epoch 2/8:  29%|██▊       | 814/2855 [32:04<58:58,  1.73s/it, loss=0.0674, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 815/2855 [32:04<58:33,  1.72s/it, loss=0.0674, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 815/2855 [32:06<58:33,  1.72s/it, loss=0.1316, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 816/2855 [32:06<56:37,  1.67s/it, loss=0.1316, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 816/2855 [32:07<56:37,  1.67s/it, loss=0.1781, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 817/2855 [32:07<55:35,  1.64s/it, loss=0.1781, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 817/2855 [32:09<55:35,  1.64s/it, loss=0.1388, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 818/2855 [32:09<56:03,  1.65s/it, loss=0.1388, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 818/2855 [32:10<56:03,  1.65s/it, loss=0.2882, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 819/2855 [32:10<54:44,  1.61s/it, loss=0.2882, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 819/2855 [32:12<54:44,  1.61s/it, loss=0.0758, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 820/2855 [32:12<52:34,  1.55s/it, loss=0.0758, lr=2.82e-05]
Epoch 2/8:  29%|██▊       | 820/2855 [32:17<52:34,  1.55s/it, loss=0.1986, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 821/2855 [32:17<1:30:08,  2.66s/it, loss=0.1986, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 821/2855 [32:18<1:30:08,  2.66s/it, loss=0.1722, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 822/2855 [32:18<1:15:53,  2.24s/it, loss=0.1722, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 822/2855 [32:20<1:15:53,  2.24s/it, loss=0.1220, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 823/2855 [32:20<1:05:29,  1.93s/it, loss=0.1220, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 823/2855 [32:21<1:05:29,  1.93s/it, loss=0.2621, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 824/2855 [32:21<58:21,  1.72s/it, loss=0.2621, lr=2.82e-05]  
Epoch 2/8:  29%|██▉       | 824/2855 [32:22<58:21,  1.72s/it, loss=0.1078, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 825/2855 [32:22<54:11,  1.60s/it, loss=0.1078, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 825/2855 [32:24<54:11,  1.60s/it, loss=0.1655, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 826/2855 [32:24<53:02,  1.57s/it, loss=0.1655, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 826/2855 [32:25<53:02,  1.57s/it, loss=0.1029, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 827/2855 [32:25<52:03,  1.54s/it, loss=0.1029, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 827/2855 [32:27<52:03,  1.54s/it, loss=0.2132, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 828/2855 [32:27<51:05,  1.51s/it, loss=0.2132, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 828/2855 [32:28<51:05,  1.51s/it, loss=0.1280, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 829/2855 [32:28<52:17,  1.55s/it, loss=0.1280, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 829/2855 [32:30<52:17,  1.55s/it, loss=0.1741, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 830/2855 [32:30<53:31,  1.59s/it, loss=0.1741, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 830/2855 [32:31<53:31,  1.59s/it, loss=0.1388, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 831/2855 [32:31<53:41,  1.59s/it, loss=0.1388, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 831/2855 [32:33<53:41,  1.59s/it, loss=0.2999, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 832/2855 [32:33<54:27,  1.62s/it, loss=0.2999, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 832/2855 [32:35<54:27,  1.62s/it, loss=0.0699, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 833/2855 [32:35<54:01,  1.60s/it, loss=0.0699, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 833/2855 [32:36<54:01,  1.60s/it, loss=0.1576, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 834/2855 [32:36<53:45,  1.60s/it, loss=0.1576, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 834/2855 [32:38<53:45,  1.60s/it, loss=0.1371, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 835/2855 [32:38<53:24,  1.59s/it, loss=0.1371, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 835/2855 [32:40<53:24,  1.59s/it, loss=0.0780, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 836/2855 [32:40<54:44,  1.63s/it, loss=0.0780, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 836/2855 [32:41<54:44,  1.63s/it, loss=0.1147, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 837/2855 [32:41<57:09,  1.70s/it, loss=0.1147, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 837/2855 [32:43<57:09,  1.70s/it, loss=0.0743, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 838/2855 [32:43<57:10,  1.70s/it, loss=0.0743, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 838/2855 [32:45<57:10,  1.70s/it, loss=0.0843, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 839/2855 [32:45<57:09,  1.70s/it, loss=0.0843, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 839/2855 [32:47<57:09,  1.70s/it, loss=0.0464, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 840/2855 [32:47<57:19,  1.71s/it, loss=0.0464, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 840/2855 [32:48<57:19,  1.71s/it, loss=0.1919, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 841/2855 [32:48<57:26,  1.71s/it, loss=0.1919, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 841/2855 [32:50<57:26,  1.71s/it, loss=0.1093, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 842/2855 [32:50<58:21,  1.74s/it, loss=0.1093, lr=2.82e-05]
Epoch 2/8:  29%|██▉       | 842/2855 [32:52<58:21,  1.74s/it, loss=0.0943, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 843/2855 [32:52<56:36,  1.69s/it, loss=0.0943, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 843/2855 [32:53<56:36,  1.69s/it, loss=0.2451, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 844/2855 [32:53<55:53,  1.67s/it, loss=0.2451, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 844/2855 [32:55<55:53,  1.67s/it, loss=0.0964, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 845/2855 [32:55<55:08,  1.65s/it, loss=0.0964, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 845/2855 [32:56<55:08,  1.65s/it, loss=0.1309, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 846/2855 [32:56<53:39,  1.60s/it, loss=0.1309, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 846/2855 [32:58<53:39,  1.60s/it, loss=0.0983, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 847/2855 [32:58<55:30,  1.66s/it, loss=0.0983, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 847/2855 [33:00<55:30,  1.66s/it, loss=0.1548, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 848/2855 [33:00<56:23,  1.69s/it, loss=0.1548, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 848/2855 [33:01<56:23,  1.69s/it, loss=0.1261, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 849/2855 [33:01<55:26,  1.66s/it, loss=0.1261, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 849/2855 [33:03<55:26,  1.66s/it, loss=0.2527, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 850/2855 [33:03<55:07,  1.65s/it, loss=0.2527, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 850/2855 [33:05<55:07,  1.65s/it, loss=0.1956, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 851/2855 [33:05<55:09,  1.65s/it, loss=0.1956, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 851/2855 [33:06<55:09,  1.65s/it, loss=0.1580, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 852/2855 [33:06<55:02,  1.65s/it, loss=0.1580, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 852/2855 [33:08<55:02,  1.65s/it, loss=0.1770, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 853/2855 [33:08<54:21,  1.63s/it, loss=0.1770, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 853/2855 [33:10<54:21,  1.63s/it, loss=0.1192, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 854/2855 [33:10<53:30,  1.60s/it, loss=0.1192, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 854/2855 [33:11<53:30,  1.60s/it, loss=0.1707, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 855/2855 [33:11<54:46,  1.64s/it, loss=0.1707, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 855/2855 [33:13<54:46,  1.64s/it, loss=0.1784, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 856/2855 [33:13<53:37,  1.61s/it, loss=0.1784, lr=2.82e-05]
Epoch 2/8:  30%|██▉       | 856/2855 [33:15<53:37,  1.61s/it, loss=0.0672, lr=2.82e-05]
Epoch 2/8:  30%|███       | 857/2855 [33:15<54:43,  1.64s/it, loss=0.0672, lr=2.82e-05]
Epoch 2/8:  30%|███       | 857/2855 [33:16<54:43,  1.64s/it, loss=0.0810, lr=2.81e-05]
Epoch 2/8:  30%|███       | 858/2855 [33:16<54:12,  1.63s/it, loss=0.0810, lr=2.81e-05]
Epoch 2/8:  30%|███       | 858/2855 [33:18<54:12,  1.63s/it, loss=0.1205, lr=2.81e-05]
Epoch 2/8:  30%|███       | 859/2855 [33:18<52:02,  1.56s/it, loss=0.1205, lr=2.81e-05]
Epoch 2/8:  30%|███       | 859/2855 [33:20<52:02,  1.56s/it, loss=0.0566, lr=2.81e-05]
Epoch 2/8:  30%|███       | 860/2855 [33:20<56:44,  1.71s/it, loss=0.0566, lr=2.81e-05]
Epoch 2/8:  30%|███       | 860/2855 [33:21<56:44,  1.71s/it, loss=0.1121, lr=2.81e-05]
Epoch 2/8:  30%|███       | 861/2855 [33:21<52:21,  1.58s/it, loss=0.1121, lr=2.81e-05]
Epoch 2/8:  30%|███       | 861/2855 [33:22<52:21,  1.58s/it, loss=0.1662, lr=2.81e-05]
Epoch 2/8:  30%|███       | 862/2855 [33:22<50:20,  1.52s/it, loss=0.1662, lr=2.81e-05]
Epoch 2/8:  30%|███       | 862/2855 [33:24<50:20,  1.52s/it, loss=0.1893, lr=2.81e-05]
Epoch 2/8:  30%|███       | 863/2855 [33:24<50:02,  1.51s/it, loss=0.1893, lr=2.81e-05]
Epoch 2/8:  30%|███       | 863/2855 [33:25<50:02,  1.51s/it, loss=0.1941, lr=2.81e-05]
Epoch 2/8:  30%|███       | 864/2855 [33:25<50:27,  1.52s/it, loss=0.1941, lr=2.81e-05]
Epoch 2/8:  30%|███       | 864/2855 [33:27<50:27,  1.52s/it, loss=0.0976, lr=2.81e-05]
Epoch 2/8:  30%|███       | 865/2855 [33:27<52:12,  1.57s/it, loss=0.0976, lr=2.81e-05]
Epoch 2/8:  30%|███       | 865/2855 [33:29<52:12,  1.57s/it, loss=0.0556, lr=2.81e-05]
Epoch 2/8:  30%|███       | 866/2855 [33:29<53:29,  1.61s/it, loss=0.0556, lr=2.81e-05]
Epoch 2/8:  30%|███       | 866/2855 [33:30<53:29,  1.61s/it, loss=0.1884, lr=2.81e-05]
Epoch 2/8:  30%|███       | 867/2855 [33:30<54:25,  1.64s/it, loss=0.1884, lr=2.81e-05]
Epoch 2/8:  30%|███       | 867/2855 [33:32<54:25,  1.64s/it, loss=0.1809, lr=2.81e-05]
Epoch 2/8:  30%|███       | 868/2855 [33:32<53:52,  1.63s/it, loss=0.1809, lr=2.81e-05]
Epoch 2/8:  30%|███       | 868/2855 [33:34<53:52,  1.63s/it, loss=0.1795, lr=2.81e-05]
Epoch 2/8:  30%|███       | 869/2855 [33:34<54:42,  1.65s/it, loss=0.1795, lr=2.81e-05]
Epoch 2/8:  30%|███       | 869/2855 [33:35<54:42,  1.65s/it, loss=0.0810, lr=2.81e-05]
Epoch 2/8:  30%|███       | 870/2855 [33:35<54:33,  1.65s/it, loss=0.0810, lr=2.81e-05]
Epoch 2/8:  30%|███       | 870/2855 [33:37<54:33,  1.65s/it, loss=0.1389, lr=2.81e-05]
Epoch 2/8:  31%|███       | 871/2855 [33:37<53:38,  1.62s/it, loss=0.1389, lr=2.81e-05]
Epoch 2/8:  31%|███       | 871/2855 [33:39<53:38,  1.62s/it, loss=0.1472, lr=2.81e-05]
Epoch 2/8:  31%|███       | 872/2855 [33:39<54:49,  1.66s/it, loss=0.1472, lr=2.81e-05]
Epoch 2/8:  31%|███       | 872/2855 [33:41<54:49,  1.66s/it, loss=0.1407, lr=2.81e-05]
Epoch 2/8:  31%|███       | 873/2855 [33:41<57:43,  1.75s/it, loss=0.1407, lr=2.81e-05]
Epoch 2/8:  31%|███       | 873/2855 [33:42<57:43,  1.75s/it, loss=0.0563, lr=2.81e-05]
Epoch 2/8:  31%|███       | 874/2855 [33:42<57:24,  1.74s/it, loss=0.0563, lr=2.81e-05]
Epoch 2/8:  31%|███       | 874/2855 [33:44<57:24,  1.74s/it, loss=0.1624, lr=2.81e-05]
Epoch 2/8:  31%|███       | 875/2855 [33:44<57:03,  1.73s/it, loss=0.1624, lr=2.81e-05]
Epoch 2/8:  31%|███       | 875/2855 [33:46<57:03,  1.73s/it, loss=0.0545, lr=2.81e-05]
Epoch 2/8:  31%|███       | 876/2855 [33:46<59:16,  1.80s/it, loss=0.0545, lr=2.81e-05]
Epoch 2/8:  31%|███       | 876/2855 [33:48<59:16,  1.80s/it, loss=0.0946, lr=2.81e-05]
Epoch 2/8:  31%|███       | 877/2855 [33:48<58:18,  1.77s/it, loss=0.0946, lr=2.81e-05]
Epoch 2/8:  31%|███       | 877/2855 [33:49<58:18,  1.77s/it, loss=0.1440, lr=2.81e-05]
Epoch 2/8:  31%|███       | 878/2855 [33:49<59:05,  1.79s/it, loss=0.1440, lr=2.81e-05]
Epoch 2/8:  31%|███       | 878/2855 [33:51<59:05,  1.79s/it, loss=0.1495, lr=2.81e-05]
Epoch 2/8:  31%|███       | 879/2855 [33:51<59:58,  1.82s/it, loss=0.1495, lr=2.81e-05]
Epoch 2/8:  31%|███       | 879/2855 [33:53<59:58,  1.82s/it, loss=0.1614, lr=2.81e-05]
Epoch 2/8:  31%|███       | 880/2855 [33:53<59:13,  1.80s/it, loss=0.1614, lr=2.81e-05]
Epoch 2/8:  31%|███       | 880/2855 [33:55<59:13,  1.80s/it, loss=0.1860, lr=2.81e-05]
Epoch 2/8:  31%|███       | 881/2855 [33:55<59:29,  1.81s/it, loss=0.1860, lr=2.81e-05]
Epoch 2/8:  31%|███       | 881/2855 [33:57<59:29,  1.81s/it, loss=0.1619, lr=2.81e-05]
Epoch 2/8:  31%|███       | 882/2855 [33:57<58:36,  1.78s/it, loss=0.1619, lr=2.81e-05]
Epoch 2/8:  31%|███       | 882/2855 [33:58<58:36,  1.78s/it, loss=0.1446, lr=2.81e-05]
Epoch 2/8:  31%|███       | 883/2855 [33:58<56:18,  1.71s/it, loss=0.1446, lr=2.81e-05]
Epoch 2/8:  31%|███       | 883/2855 [34:00<56:18,  1.71s/it, loss=0.0902, lr=2.81e-05]
Epoch 2/8:  31%|███       | 884/2855 [34:00<55:09,  1.68s/it, loss=0.0902, lr=2.81e-05]
Epoch 2/8:  31%|███       | 884/2855 [34:01<55:09,  1.68s/it, loss=0.1623, lr=2.81e-05]
Epoch 2/8:  31%|███       | 885/2855 [34:01<54:26,  1.66s/it, loss=0.1623, lr=2.81e-05]
Epoch 2/8:  31%|███       | 885/2855 [34:03<54:26,  1.66s/it, loss=0.0041, lr=2.81e-05]
Epoch 2/8:  31%|███       | 886/2855 [34:03<52:54,  1.61s/it, loss=0.0041, lr=2.81e-05]
Epoch 2/8:  31%|███       | 886/2855 [34:05<52:54,  1.61s/it, loss=0.2009, lr=2.81e-05]
Epoch 2/8:  31%|███       | 887/2855 [34:05<56:03,  1.71s/it, loss=0.2009, lr=2.81e-05]
Epoch 2/8:  31%|███       | 887/2855 [34:07<56:03,  1.71s/it, loss=0.2775, lr=2.81e-05]
Epoch 2/8:  31%|███       | 888/2855 [34:07<57:33,  1.76s/it, loss=0.2775, lr=2.81e-05]
Epoch 2/8:  31%|███       | 888/2855 [34:09<57:33,  1.76s/it, loss=0.2659, lr=2.81e-05]
Epoch 2/8:  31%|███       | 889/2855 [34:09<59:59,  1.83s/it, loss=0.2659, lr=2.81e-05]
Epoch 2/8:  31%|███       | 889/2855 [34:10<59:59,  1.83s/it, loss=0.1669, lr=2.81e-05]
Epoch 2/8:  31%|███       | 890/2855 [34:10<59:05,  1.80s/it, loss=0.1669, lr=2.81e-05]
Epoch 2/8:  31%|███       | 890/2855 [34:12<59:05,  1.80s/it, loss=0.1615, lr=2.81e-05]
Epoch 2/8:  31%|███       | 891/2855 [34:12<57:19,  1.75s/it, loss=0.1615, lr=2.81e-05]
Epoch 2/8:  31%|███       | 891/2855 [34:14<57:19,  1.75s/it, loss=0.1732, lr=2.81e-05]
Epoch 2/8:  31%|███       | 892/2855 [34:14<57:16,  1.75s/it, loss=0.1732, lr=2.81e-05]
Epoch 2/8:  31%|███       | 892/2855 [34:16<57:16,  1.75s/it, loss=0.0722, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 893/2855 [34:16<59:15,  1.81s/it, loss=0.0722, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 893/2855 [34:18<59:15,  1.81s/it, loss=0.1834, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 894/2855 [34:18<59:46,  1.83s/it, loss=0.1834, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 894/2855 [34:20<59:46,  1.83s/it, loss=0.2682, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 895/2855 [34:20<1:02:11,  1.90s/it, loss=0.2682, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 895/2855 [34:22<1:02:11,  1.90s/it, loss=0.1911, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 896/2855 [34:22<1:00:34,  1.86s/it, loss=0.1911, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 896/2855 [34:23<1:00:34,  1.86s/it, loss=0.0385, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 897/2855 [34:23<59:17,  1.82s/it, loss=0.0385, lr=2.81e-05]  
Epoch 2/8:  31%|███▏      | 897/2855 [34:25<59:17,  1.82s/it, loss=0.0520, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 898/2855 [34:25<59:46,  1.83s/it, loss=0.0520, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 898/2855 [34:27<59:46,  1.83s/it, loss=0.2043, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 899/2855 [34:27<58:11,  1.78s/it, loss=0.2043, lr=2.81e-05]
Epoch 2/8:  31%|███▏      | 899/2855 [34:33<58:11,  1.78s/it, loss=0.0570, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 900/2855 [34:33<1:36:45,  2.97s/it, loss=0.0570, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 900/2855 [34:34<1:36:45,  2.97s/it, loss=0.0441, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 901/2855 [34:34<1:25:36,  2.63s/it, loss=0.0441, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 901/2855 [34:37<1:25:36,  2.63s/it, loss=0.1764, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 902/2855 [34:37<1:21:14,  2.50s/it, loss=0.1764, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 902/2855 [34:38<1:21:14,  2.50s/it, loss=0.1889, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 903/2855 [34:38<1:15:25,  2.32s/it, loss=0.1889, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 903/2855 [34:40<1:15:25,  2.32s/it, loss=0.0913, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 904/2855 [34:40<1:11:51,  2.21s/it, loss=0.0913, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 904/2855 [34:42<1:11:51,  2.21s/it, loss=0.0918, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 905/2855 [34:42<1:07:35,  2.08s/it, loss=0.0918, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 905/2855 [34:44<1:07:35,  2.08s/it, loss=0.2065, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 906/2855 [34:44<1:05:40,  2.02s/it, loss=0.2065, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 906/2855 [34:46<1:05:40,  2.02s/it, loss=0.2525, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 907/2855 [34:46<1:04:47,  2.00s/it, loss=0.2525, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 907/2855 [34:48<1:04:47,  2.00s/it, loss=0.0997, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 908/2855 [34:48<1:03:26,  1.96s/it, loss=0.0997, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 908/2855 [34:50<1:03:26,  1.96s/it, loss=0.0935, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 909/2855 [34:50<1:09:00,  2.13s/it, loss=0.0935, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 909/2855 [34:53<1:09:00,  2.13s/it, loss=0.1655, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 910/2855 [34:53<1:10:13,  2.17s/it, loss=0.1655, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 910/2855 [34:55<1:10:13,  2.17s/it, loss=0.0355, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 911/2855 [34:55<1:12:32,  2.24s/it, loss=0.0355, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 911/2855 [34:57<1:12:32,  2.24s/it, loss=0.1099, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 912/2855 [34:57<1:08:06,  2.10s/it, loss=0.1099, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 912/2855 [34:59<1:08:06,  2.10s/it, loss=0.2656, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 913/2855 [34:59<1:05:56,  2.04s/it, loss=0.2656, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 913/2855 [35:01<1:05:56,  2.04s/it, loss=0.2005, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 914/2855 [35:01<1:06:41,  2.06s/it, loss=0.2005, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 914/2855 [35:03<1:06:41,  2.06s/it, loss=0.1448, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 915/2855 [35:03<1:06:11,  2.05s/it, loss=0.1448, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 915/2855 [35:05<1:06:11,  2.05s/it, loss=0.1879, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 916/2855 [35:05<1:05:03,  2.01s/it, loss=0.1879, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 916/2855 [35:07<1:05:03,  2.01s/it, loss=0.0678, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 917/2855 [35:07<1:03:19,  1.96s/it, loss=0.0678, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 917/2855 [35:08<1:03:19,  1.96s/it, loss=0.0383, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 918/2855 [35:08<1:00:58,  1.89s/it, loss=0.0383, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 918/2855 [35:10<1:00:58,  1.89s/it, loss=0.2072, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 919/2855 [35:10<1:00:28,  1.87s/it, loss=0.2072, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 919/2855 [35:12<1:00:28,  1.87s/it, loss=0.0961, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 920/2855 [35:12<58:32,  1.82s/it, loss=0.0961, lr=2.81e-05]  
Epoch 2/8:  32%|███▏      | 920/2855 [35:14<58:32,  1.82s/it, loss=0.1101, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 921/2855 [35:14<57:50,  1.79s/it, loss=0.1101, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 921/2855 [35:16<57:50,  1.79s/it, loss=0.2444, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 922/2855 [35:16<58:50,  1.83s/it, loss=0.2444, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 922/2855 [35:17<58:50,  1.83s/it, loss=0.1278, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 923/2855 [35:17<58:33,  1.82s/it, loss=0.1278, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 923/2855 [35:19<58:33,  1.82s/it, loss=0.2693, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 924/2855 [35:19<58:53,  1.83s/it, loss=0.2693, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 924/2855 [35:21<58:53,  1.83s/it, loss=0.0001, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 925/2855 [35:21<1:00:37,  1.88s/it, loss=0.0001, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 925/2855 [35:23<1:00:37,  1.88s/it, loss=0.0439, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 926/2855 [35:23<1:00:16,  1.87s/it, loss=0.0439, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 926/2855 [35:25<1:00:16,  1.87s/it, loss=0.1440, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 927/2855 [35:25<1:00:53,  1.89s/it, loss=0.1440, lr=2.81e-05]
Epoch 2/8:  32%|███▏      | 927/2855 [35:27<1:00:53,  1.89s/it, loss=0.1279, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 928/2855 [35:27<1:02:24,  1.94s/it, loss=0.1279, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 928/2855 [35:29<1:02:24,  1.94s/it, loss=0.1941, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 929/2855 [35:29<1:01:46,  1.92s/it, loss=0.1941, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 929/2855 [35:31<1:01:46,  1.92s/it, loss=0.1032, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 930/2855 [35:31<1:03:36,  1.98s/it, loss=0.1032, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 930/2855 [35:33<1:03:36,  1.98s/it, loss=0.1212, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 931/2855 [35:33<1:04:57,  2.03s/it, loss=0.1212, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 931/2855 [35:35<1:04:57,  2.03s/it, loss=0.1398, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 932/2855 [35:35<1:06:09,  2.06s/it, loss=0.1398, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 932/2855 [35:37<1:06:09,  2.06s/it, loss=0.1617, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 933/2855 [35:37<1:04:48,  2.02s/it, loss=0.1617, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 933/2855 [35:41<1:04:48,  2.02s/it, loss=0.2352, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 934/2855 [35:41<1:18:57,  2.47s/it, loss=0.2352, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 934/2855 [35:43<1:18:57,  2.47s/it, loss=0.0961, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 935/2855 [35:43<1:12:24,  2.26s/it, loss=0.0961, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 935/2855 [35:45<1:12:24,  2.26s/it, loss=0.1750, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 936/2855 [35:45<1:10:34,  2.21s/it, loss=0.1750, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 936/2855 [35:46<1:10:34,  2.21s/it, loss=0.1715, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 937/2855 [35:46<1:07:28,  2.11s/it, loss=0.1715, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 937/2855 [35:48<1:07:28,  2.11s/it, loss=0.1093, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 938/2855 [35:48<1:04:26,  2.02s/it, loss=0.1093, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 938/2855 [35:54<1:04:26,  2.02s/it, loss=0.1045, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 939/2855 [35:54<1:38:14,  3.08s/it, loss=0.1045, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 939/2855 [35:55<1:38:14,  3.08s/it, loss=0.2872, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 940/2855 [35:55<1:22:26,  2.58s/it, loss=0.2872, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 940/2855 [35:57<1:22:26,  2.58s/it, loss=0.0751, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 941/2855 [35:57<1:15:01,  2.35s/it, loss=0.0751, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 941/2855 [35:59<1:15:01,  2.35s/it, loss=0.1110, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 942/2855 [35:59<1:12:34,  2.28s/it, loss=0.1110, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 942/2855 [36:01<1:12:34,  2.28s/it, loss=0.0900, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 943/2855 [36:01<1:08:39,  2.15s/it, loss=0.0900, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 943/2855 [36:03<1:08:39,  2.15s/it, loss=0.0374, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 944/2855 [36:03<1:04:52,  2.04s/it, loss=0.0374, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 944/2855 [36:05<1:04:52,  2.04s/it, loss=0.1280, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 945/2855 [36:05<1:04:47,  2.04s/it, loss=0.1280, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 945/2855 [36:07<1:04:47,  2.04s/it, loss=0.2096, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 946/2855 [36:07<1:03:24,  1.99s/it, loss=0.2096, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 946/2855 [36:09<1:03:24,  1.99s/it, loss=0.0777, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 947/2855 [36:09<1:01:26,  1.93s/it, loss=0.0777, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 947/2855 [36:10<1:01:26,  1.93s/it, loss=0.0128, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 948/2855 [36:10<1:01:11,  1.93s/it, loss=0.0128, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 948/2855 [36:12<1:01:11,  1.93s/it, loss=0.1714, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 949/2855 [36:12<59:50,  1.88s/it, loss=0.1714, lr=2.81e-05]  
Epoch 2/8:  33%|███▎      | 949/2855 [36:14<59:50,  1.88s/it, loss=0.1174, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 950/2855 [36:14<59:04,  1.86s/it, loss=0.1174, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 950/2855 [36:16<59:04,  1.86s/it, loss=0.0839, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 951/2855 [36:16<1:00:23,  1.90s/it, loss=0.0839, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 951/2855 [36:18<1:00:23,  1.90s/it, loss=0.1623, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 952/2855 [36:18<59:00,  1.86s/it, loss=0.1623, lr=2.81e-05]  
Epoch 2/8:  33%|███▎      | 952/2855 [36:20<59:00,  1.86s/it, loss=0.1012, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 953/2855 [36:20<57:50,  1.82s/it, loss=0.1012, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 953/2855 [36:21<57:50,  1.82s/it, loss=0.1664, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 954/2855 [36:21<56:44,  1.79s/it, loss=0.1664, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 954/2855 [36:23<56:44,  1.79s/it, loss=0.0652, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 955/2855 [36:23<54:33,  1.72s/it, loss=0.0652, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 955/2855 [36:24<54:33,  1.72s/it, loss=0.2691, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 956/2855 [36:24<53:34,  1.69s/it, loss=0.2691, lr=2.81e-05]
Epoch 2/8:  33%|███▎      | 956/2855 [36:26<53:34,  1.69s/it, loss=0.1424, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 957/2855 [36:26<53:51,  1.70s/it, loss=0.1424, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 957/2855 [36:28<53:51,  1.70s/it, loss=0.1182, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 958/2855 [36:28<53:03,  1.68s/it, loss=0.1182, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 958/2855 [36:30<53:03,  1.68s/it, loss=0.1720, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 959/2855 [36:30<56:33,  1.79s/it, loss=0.1720, lr=2.81e-05]
Epoch 2/8:  34%|███▎      | 959/2855 [36:32<56:33,  1.79s/it, loss=0.2223, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 960/2855 [36:32<59:10,  1.87s/it, loss=0.2223, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 960/2855 [36:34<59:10,  1.87s/it, loss=0.0766, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 961/2855 [36:34<57:19,  1.82s/it, loss=0.0766, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 961/2855 [36:36<57:19,  1.82s/it, loss=0.2234, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 962/2855 [36:36<58:46,  1.86s/it, loss=0.2234, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 962/2855 [36:37<58:46,  1.86s/it, loss=0.1240, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 963/2855 [36:37<58:08,  1.84s/it, loss=0.1240, lr=2.80e-05]
Epoch 2/8:  34%|███▎      | 963/2855 [36:39<58:08,  1.84s/it, loss=0.2034, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 964/2855 [36:39<56:43,  1.80s/it, loss=0.2034, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 964/2855 [36:41<56:43,  1.80s/it, loss=0.1462, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 965/2855 [36:41<57:50,  1.84s/it, loss=0.1462, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 965/2855 [36:43<57:50,  1.84s/it, loss=0.1672, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 966/2855 [36:43<57:37,  1.83s/it, loss=0.1672, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 966/2855 [36:45<57:37,  1.83s/it, loss=0.1714, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 967/2855 [36:45<57:22,  1.82s/it, loss=0.1714, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 967/2855 [36:47<57:22,  1.82s/it, loss=0.2113, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 968/2855 [36:47<58:43,  1.87s/it, loss=0.2113, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 968/2855 [36:48<58:43,  1.87s/it, loss=0.1436, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 969/2855 [36:48<58:09,  1.85s/it, loss=0.1436, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 969/2855 [36:50<58:09,  1.85s/it, loss=0.0897, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 970/2855 [36:50<59:33,  1.90s/it, loss=0.0897, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 970/2855 [36:52<59:33,  1.90s/it, loss=0.1306, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 971/2855 [36:52<59:12,  1.89s/it, loss=0.1306, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 971/2855 [36:54<59:12,  1.89s/it, loss=0.3123, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 972/2855 [36:54<58:17,  1.86s/it, loss=0.3123, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 972/2855 [36:56<58:17,  1.86s/it, loss=0.2148, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 973/2855 [36:56<58:06,  1.85s/it, loss=0.2148, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 973/2855 [36:57<58:06,  1.85s/it, loss=0.1593, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 974/2855 [36:57<55:29,  1.77s/it, loss=0.1593, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 974/2855 [36:59<55:29,  1.77s/it, loss=0.1532, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 975/2855 [36:59<53:15,  1.70s/it, loss=0.1532, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 975/2855 [37:01<53:15,  1.70s/it, loss=0.1671, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 976/2855 [37:01<57:56,  1.85s/it, loss=0.1671, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 976/2855 [37:03<57:56,  1.85s/it, loss=0.1157, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 977/2855 [37:03<59:33,  1.90s/it, loss=0.1157, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 977/2855 [37:05<59:33,  1.90s/it, loss=0.0763, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 978/2855 [37:05<59:27,  1.90s/it, loss=0.0763, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 978/2855 [37:11<59:27,  1.90s/it, loss=0.1437, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 979/2855 [37:11<1:33:42,  3.00s/it, loss=0.1437, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 979/2855 [37:12<1:33:42,  3.00s/it, loss=0.1367, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 980/2855 [37:12<1:18:51,  2.52s/it, loss=0.1367, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 980/2855 [37:14<1:18:51,  2.52s/it, loss=0.1232, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 981/2855 [37:14<1:13:56,  2.37s/it, loss=0.1232, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 981/2855 [37:16<1:13:56,  2.37s/it, loss=0.0899, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 982/2855 [37:16<1:13:18,  2.35s/it, loss=0.0899, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 982/2855 [37:18<1:13:18,  2.35s/it, loss=0.1154, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 983/2855 [37:18<1:07:34,  2.17s/it, loss=0.1154, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 983/2855 [37:20<1:07:34,  2.17s/it, loss=0.2931, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 984/2855 [37:20<1:05:35,  2.10s/it, loss=0.2931, lr=2.80e-05]
Epoch 2/8:  34%|███▍      | 984/2855 [37:22<1:05:35,  2.10s/it, loss=0.2307, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 985/2855 [37:22<1:05:56,  2.12s/it, loss=0.2307, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 985/2855 [37:24<1:05:56,  2.12s/it, loss=0.2361, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 986/2855 [37:24<1:03:31,  2.04s/it, loss=0.2361, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 986/2855 [37:26<1:03:31,  2.04s/it, loss=0.1404, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 987/2855 [37:26<1:02:54,  2.02s/it, loss=0.1404, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 987/2855 [37:28<1:02:54,  2.02s/it, loss=0.2219, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 988/2855 [37:28<1:00:17,  1.94s/it, loss=0.2219, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 988/2855 [37:30<1:00:17,  1.94s/it, loss=0.2774, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 989/2855 [37:30<1:00:34,  1.95s/it, loss=0.2774, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 989/2855 [37:32<1:00:34,  1.95s/it, loss=0.1231, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 990/2855 [37:32<1:04:23,  2.07s/it, loss=0.1231, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 990/2855 [37:34<1:04:23,  2.07s/it, loss=0.2261, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 991/2855 [37:34<1:01:30,  1.98s/it, loss=0.2261, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 991/2855 [37:36<1:01:30,  1.98s/it, loss=0.1055, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 992/2855 [37:36<1:02:20,  2.01s/it, loss=0.1055, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 992/2855 [37:38<1:02:20,  2.01s/it, loss=0.2033, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 993/2855 [37:38<1:04:15,  2.07s/it, loss=0.2033, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 993/2855 [37:40<1:04:15,  2.07s/it, loss=0.1313, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 994/2855 [37:40<1:00:31,  1.95s/it, loss=0.1313, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 994/2855 [37:42<1:00:31,  1.95s/it, loss=0.1617, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 995/2855 [37:42<1:03:32,  2.05s/it, loss=0.1617, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 995/2855 [37:44<1:03:32,  2.05s/it, loss=0.1556, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 996/2855 [37:44<1:03:10,  2.04s/it, loss=0.1556, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 996/2855 [37:46<1:03:10,  2.04s/it, loss=0.2211, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 997/2855 [37:46<1:02:48,  2.03s/it, loss=0.2211, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 997/2855 [37:48<1:02:48,  2.03s/it, loss=0.2081, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 998/2855 [37:48<1:01:32,  1.99s/it, loss=0.2081, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 998/2855 [37:50<1:01:32,  1.99s/it, loss=0.0923, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 999/2855 [37:50<1:02:55,  2.03s/it, loss=0.0923, lr=2.80e-05]
Epoch 2/8:  35%|███▍      | 999/2855 [37:52<1:02:55,  2.03s/it, loss=0.0497, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1000/2855 [37:52<1:03:20,  2.05s/it, loss=0.0497, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1000/2855 [37:54<1:03:20,  2.05s/it, loss=0.1056, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1001/2855 [37:54<1:02:42,  2.03s/it, loss=0.1056, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1001/2855 [37:57<1:02:42,  2.03s/it, loss=0.1575, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1002/2855 [37:57<1:06:17,  2.15s/it, loss=0.1575, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1002/2855 [37:59<1:06:17,  2.15s/it, loss=0.1884, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1003/2855 [37:59<1:04:02,  2.07s/it, loss=0.1884, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1003/2855 [38:01<1:04:02,  2.07s/it, loss=0.1683, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1004/2855 [38:01<1:03:19,  2.05s/it, loss=0.1683, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1004/2855 [38:03<1:03:19,  2.05s/it, loss=0.1564, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1005/2855 [38:03<1:05:27,  2.12s/it, loss=0.1564, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1005/2855 [38:05<1:05:27,  2.12s/it, loss=0.1237, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1006/2855 [38:05<1:04:25,  2.09s/it, loss=0.1237, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1006/2855 [38:07<1:04:25,  2.09s/it, loss=0.1669, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1007/2855 [38:07<1:06:11,  2.15s/it, loss=0.1669, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1007/2855 [38:09<1:06:11,  2.15s/it, loss=0.0952, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1008/2855 [38:09<1:04:26,  2.09s/it, loss=0.0952, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1008/2855 [38:11<1:04:26,  2.09s/it, loss=0.2494, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1009/2855 [38:11<1:05:53,  2.14s/it, loss=0.2494, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1009/2855 [38:13<1:05:53,  2.14s/it, loss=0.0513, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1010/2855 [38:13<1:05:07,  2.12s/it, loss=0.0513, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1010/2855 [38:16<1:05:07,  2.12s/it, loss=0.0956, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1011/2855 [38:16<1:05:51,  2.14s/it, loss=0.0956, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1011/2855 [38:18<1:05:51,  2.14s/it, loss=0.0828, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1012/2855 [38:18<1:07:39,  2.20s/it, loss=0.0828, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1012/2855 [38:20<1:07:39,  2.20s/it, loss=0.1629, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1013/2855 [38:20<1:06:35,  2.17s/it, loss=0.1629, lr=2.80e-05]
Epoch 2/8:  35%|███▌      | 1013/2855 [38:22<1:06:35,  2.17s/it, loss=0.1330, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1014/2855 [38:22<1:06:07,  2.16s/it, loss=0.1330, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1014/2855 [38:24<1:06:07,  2.16s/it, loss=0.0884, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1015/2855 [38:24<1:05:24,  2.13s/it, loss=0.0884, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1015/2855 [38:27<1:05:24,  2.13s/it, loss=0.1155, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1016/2855 [38:27<1:06:33,  2.17s/it, loss=0.1155, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1016/2855 [38:29<1:06:33,  2.17s/it, loss=0.0297, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1017/2855 [38:29<1:06:07,  2.16s/it, loss=0.0297, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1017/2855 [38:31<1:06:07,  2.16s/it, loss=0.1377, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1018/2855 [38:31<1:04:55,  2.12s/it, loss=0.1377, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1018/2855 [38:38<1:04:55,  2.12s/it, loss=0.2535, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1019/2855 [38:38<1:49:22,  3.57s/it, loss=0.2535, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1019/2855 [38:39<1:49:22,  3.57s/it, loss=0.1249, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1020/2855 [38:39<1:32:50,  3.04s/it, loss=0.1249, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1020/2855 [38:43<1:32:50,  3.04s/it, loss=0.0534, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1021/2855 [38:43<1:34:19,  3.09s/it, loss=0.0534, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1021/2855 [38:45<1:34:19,  3.09s/it, loss=0.1508, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1022/2855 [38:45<1:24:57,  2.78s/it, loss=0.1508, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1022/2855 [38:48<1:24:57,  2.78s/it, loss=0.1591, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1023/2855 [38:48<1:26:30,  2.83s/it, loss=0.1591, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1023/2855 [38:50<1:26:30,  2.83s/it, loss=0.2369, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1024/2855 [38:50<1:22:00,  2.69s/it, loss=0.2369, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1024/2855 [38:52<1:22:00,  2.69s/it, loss=0.1805, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1025/2855 [38:52<1:18:18,  2.57s/it, loss=0.1805, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1025/2855 [38:55<1:18:18,  2.57s/it, loss=0.2351, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1026/2855 [38:55<1:15:31,  2.48s/it, loss=0.2351, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1026/2855 [38:57<1:15:31,  2.48s/it, loss=0.1366, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1027/2855 [38:57<1:15:01,  2.46s/it, loss=0.1366, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1027/2855 [38:59<1:15:01,  2.46s/it, loss=0.0740, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1028/2855 [38:59<1:11:45,  2.36s/it, loss=0.0740, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1028/2855 [39:01<1:11:45,  2.36s/it, loss=0.1870, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1029/2855 [39:01<1:11:28,  2.35s/it, loss=0.1870, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1029/2855 [39:04<1:11:28,  2.35s/it, loss=0.0792, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1030/2855 [39:04<1:11:42,  2.36s/it, loss=0.0792, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1030/2855 [39:06<1:11:42,  2.36s/it, loss=0.1972, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1031/2855 [39:06<1:11:34,  2.35s/it, loss=0.1972, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1031/2855 [39:09<1:11:34,  2.35s/it, loss=0.2593, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1032/2855 [39:09<1:13:09,  2.41s/it, loss=0.2593, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1032/2855 [39:11<1:13:09,  2.41s/it, loss=0.3206, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1033/2855 [39:11<1:10:56,  2.34s/it, loss=0.3206, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1033/2855 [39:14<1:10:56,  2.34s/it, loss=0.1197, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1034/2855 [39:14<1:13:24,  2.42s/it, loss=0.1197, lr=2.80e-05]
Epoch 2/8:  36%|███▌      | 1034/2855 [39:16<1:13:24,  2.42s/it, loss=0.0494, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1035/2855 [39:16<1:11:02,  2.34s/it, loss=0.0494, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1035/2855 [39:18<1:11:02,  2.34s/it, loss=0.1788, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1036/2855 [39:18<1:13:58,  2.44s/it, loss=0.1788, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1036/2855 [39:21<1:13:58,  2.44s/it, loss=0.1313, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1037/2855 [39:21<1:13:01,  2.41s/it, loss=0.1313, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1037/2855 [39:23<1:13:01,  2.41s/it, loss=0.1502, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1038/2855 [39:23<1:15:22,  2.49s/it, loss=0.1502, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1038/2855 [39:26<1:15:22,  2.49s/it, loss=0.0056, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1039/2855 [39:26<1:13:11,  2.42s/it, loss=0.0056, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1039/2855 [39:28<1:13:11,  2.42s/it, loss=0.1844, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1040/2855 [39:28<1:13:14,  2.42s/it, loss=0.1844, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1040/2855 [39:30<1:13:14,  2.42s/it, loss=0.0949, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1041/2855 [39:30<1:12:12,  2.39s/it, loss=0.0949, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1041/2855 [39:33<1:12:12,  2.39s/it, loss=0.2127, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1042/2855 [39:33<1:13:08,  2.42s/it, loss=0.2127, lr=2.80e-05]
Epoch 2/8:  36%|███▋      | 1042/2855 [39:35<1:13:08,  2.42s/it, loss=0.2447, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1043/2855 [39:35<1:09:14,  2.29s/it, loss=0.2447, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1043/2855 [39:37<1:09:14,  2.29s/it, loss=0.0891, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1044/2855 [39:37<1:08:48,  2.28s/it, loss=0.0891, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1044/2855 [39:39<1:08:48,  2.28s/it, loss=0.0929, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1045/2855 [39:39<1:08:18,  2.26s/it, loss=0.0929, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1045/2855 [39:41<1:08:18,  2.26s/it, loss=0.0327, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1046/2855 [39:41<1:05:37,  2.18s/it, loss=0.0327, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1046/2855 [39:44<1:05:37,  2.18s/it, loss=0.0731, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1047/2855 [39:44<1:11:12,  2.36s/it, loss=0.0731, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1047/2855 [39:46<1:11:12,  2.36s/it, loss=0.1620, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1048/2855 [39:46<1:08:41,  2.28s/it, loss=0.1620, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1048/2855 [39:49<1:08:41,  2.28s/it, loss=0.1277, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1049/2855 [39:49<1:10:41,  2.35s/it, loss=0.1277, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1049/2855 [39:51<1:10:41,  2.35s/it, loss=0.1196, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1050/2855 [39:51<1:08:33,  2.28s/it, loss=0.1196, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1050/2855 [39:53<1:08:33,  2.28s/it, loss=0.0878, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1051/2855 [39:53<1:09:52,  2.32s/it, loss=0.0878, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1051/2855 [39:55<1:09:52,  2.32s/it, loss=0.0919, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1052/2855 [39:55<1:07:14,  2.24s/it, loss=0.0919, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1052/2855 [39:57<1:07:14,  2.24s/it, loss=0.0825, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1053/2855 [39:57<1:05:14,  2.17s/it, loss=0.0825, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1053/2855 [39:59<1:05:14,  2.17s/it, loss=0.1034, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1054/2855 [39:59<1:05:19,  2.18s/it, loss=0.1034, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1054/2855 [40:02<1:05:19,  2.18s/it, loss=0.1557, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1055/2855 [40:02<1:10:46,  2.36s/it, loss=0.1557, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1055/2855 [40:05<1:10:46,  2.36s/it, loss=0.1007, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1056/2855 [40:05<1:09:46,  2.33s/it, loss=0.1007, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1056/2855 [40:07<1:09:46,  2.33s/it, loss=0.2285, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1057/2855 [40:07<1:06:43,  2.23s/it, loss=0.2285, lr=2.80e-05]
Epoch 2/8:  37%|███▋      | 1057/2855 [40:13<1:06:43,  2.23s/it, loss=0.1433, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1058/2855 [40:13<1:48:26,  3.62s/it, loss=0.1433, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1058/2855 [40:15<1:48:26,  3.62s/it, loss=0.1049, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1059/2855 [40:15<1:30:15,  3.02s/it, loss=0.1049, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1059/2855 [40:18<1:30:15,  3.02s/it, loss=0.1398, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1060/2855 [40:18<1:32:35,  3.09s/it, loss=0.1398, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1060/2855 [40:20<1:32:35,  3.09s/it, loss=0.1981, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1061/2855 [40:20<1:22:36,  2.76s/it, loss=0.1981, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1061/2855 [40:23<1:22:36,  2.76s/it, loss=0.1796, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1062/2855 [40:23<1:21:53,  2.74s/it, loss=0.1796, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1062/2855 [40:25<1:21:53,  2.74s/it, loss=0.1177, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1063/2855 [40:25<1:16:47,  2.57s/it, loss=0.1177, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1063/2855 [40:27<1:16:47,  2.57s/it, loss=0.1412, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1064/2855 [40:27<1:14:27,  2.49s/it, loss=0.1412, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1064/2855 [40:30<1:14:27,  2.49s/it, loss=0.1391, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1065/2855 [40:30<1:11:17,  2.39s/it, loss=0.1391, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1065/2855 [40:32<1:11:17,  2.39s/it, loss=0.0420, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1066/2855 [40:32<1:08:15,  2.29s/it, loss=0.0420, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1066/2855 [40:33<1:08:15,  2.29s/it, loss=0.0792, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1067/2855 [40:33<1:04:05,  2.15s/it, loss=0.0792, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1067/2855 [40:35<1:04:05,  2.15s/it, loss=0.0900, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1068/2855 [40:35<59:29,  2.00s/it, loss=0.0900, lr=2.79e-05]  
Epoch 2/8:  37%|███▋      | 1068/2855 [40:38<59:29,  2.00s/it, loss=0.2733, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1069/2855 [40:38<1:06:47,  2.24s/it, loss=0.2733, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1069/2855 [40:40<1:06:47,  2.24s/it, loss=0.0496, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1070/2855 [40:40<1:08:27,  2.30s/it, loss=0.0496, lr=2.79e-05]
Epoch 2/8:  37%|███▋      | 1070/2855 [40:43<1:08:27,  2.30s/it, loss=0.0677, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1071/2855 [40:43<1:08:35,  2.31s/it, loss=0.0677, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1071/2855 [40:45<1:08:35,  2.31s/it, loss=0.1888, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1072/2855 [40:45<1:07:16,  2.26s/it, loss=0.1888, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1072/2855 [40:47<1:07:16,  2.26s/it, loss=0.0987, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1073/2855 [40:47<1:06:56,  2.25s/it, loss=0.0987, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1073/2855 [40:49<1:06:56,  2.25s/it, loss=0.1510, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1074/2855 [40:49<1:05:26,  2.20s/it, loss=0.1510, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1074/2855 [40:51<1:05:26,  2.20s/it, loss=0.0561, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1075/2855 [40:51<1:06:13,  2.23s/it, loss=0.0561, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1075/2855 [40:54<1:06:13,  2.23s/it, loss=0.0224, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1076/2855 [40:54<1:11:56,  2.43s/it, loss=0.0224, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1076/2855 [40:56<1:11:56,  2.43s/it, loss=0.1118, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1077/2855 [40:56<1:07:49,  2.29s/it, loss=0.1118, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1077/2855 [40:59<1:07:49,  2.29s/it, loss=0.1883, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1078/2855 [40:59<1:07:22,  2.27s/it, loss=0.1883, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1078/2855 [41:01<1:07:22,  2.27s/it, loss=0.1395, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1079/2855 [41:01<1:05:19,  2.21s/it, loss=0.1395, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1079/2855 [41:03<1:05:19,  2.21s/it, loss=0.2247, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1080/2855 [41:03<1:07:23,  2.28s/it, loss=0.2247, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1080/2855 [41:05<1:07:23,  2.28s/it, loss=0.0713, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1081/2855 [41:05<1:05:17,  2.21s/it, loss=0.0713, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1081/2855 [41:08<1:05:17,  2.21s/it, loss=0.1457, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1082/2855 [41:08<1:07:32,  2.29s/it, loss=0.1457, lr=2.79e-05]
Epoch 2/8:  38%|██���▊      | 1082/2855 [41:10<1:07:32,  2.29s/it, loss=0.2495, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1083/2855 [41:10<1:08:54,  2.33s/it, loss=0.2495, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1083/2855 [41:12<1:08:54,  2.33s/it, loss=0.2353, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1084/2855 [41:12<1:08:34,  2.32s/it, loss=0.2353, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1084/2855 [41:15<1:08:34,  2.32s/it, loss=0.1751, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1085/2855 [41:15<1:08:24,  2.32s/it, loss=0.1751, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1085/2855 [41:17<1:08:24,  2.32s/it, loss=0.2305, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1086/2855 [41:17<1:06:39,  2.26s/it, loss=0.2305, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1086/2855 [41:20<1:06:39,  2.26s/it, loss=0.1149, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1087/2855 [41:20<1:11:13,  2.42s/it, loss=0.1149, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1087/2855 [41:22<1:11:13,  2.42s/it, loss=0.1083, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1088/2855 [41:22<1:10:06,  2.38s/it, loss=0.1083, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1088/2855 [41:24<1:10:06,  2.38s/it, loss=0.1909, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1089/2855 [41:24<1:10:35,  2.40s/it, loss=0.1909, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1089/2855 [41:26<1:10:35,  2.40s/it, loss=0.2090, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1090/2855 [41:26<1:09:14,  2.35s/it, loss=0.2090, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1090/2855 [41:29<1:09:14,  2.35s/it, loss=0.2170, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1091/2855 [41:29<1:10:07,  2.39s/it, loss=0.2170, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1091/2855 [41:31<1:10:07,  2.39s/it, loss=0.0720, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1092/2855 [41:31<1:10:08,  2.39s/it, loss=0.0720, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1092/2855 [41:34<1:10:08,  2.39s/it, loss=0.2833, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1093/2855 [41:34<1:12:08,  2.46s/it, loss=0.2833, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1093/2855 [41:36<1:12:08,  2.46s/it, loss=0.1371, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1094/2855 [41:36<1:09:50,  2.38s/it, loss=0.1371, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1094/2855 [41:39<1:09:50,  2.38s/it, loss=0.1753, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1095/2855 [41:39<1:10:56,  2.42s/it, loss=0.1753, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1095/2855 [41:41<1:10:56,  2.42s/it, loss=0.1344, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1096/2855 [41:41<1:10:19,  2.40s/it, loss=0.1344, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1096/2855 [41:43<1:10:19,  2.40s/it, loss=0.0473, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1097/2855 [41:43<1:10:54,  2.42s/it, loss=0.0473, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1097/2855 [41:49<1:10:54,  2.42s/it, loss=0.1815, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1098/2855 [41:49<1:37:17,  3.32s/it, loss=0.1815, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1098/2855 [41:52<1:37:17,  3.32s/it, loss=0.2799, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1099/2855 [41:52<1:33:38,  3.20s/it, loss=0.2799, lr=2.79e-05]
Epoch 2/8:  38%|███▊      | 1099/2855 [41:54<1:33:38,  3.20s/it, loss=0.1654, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1100/2855 [41:54<1:26:48,  2.97s/it, loss=0.1654, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1100/2855 [41:57<1:26:48,  2.97s/it, loss=0.2037, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1101/2855 [41:57<1:20:46,  2.76s/it, loss=0.2037, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1101/2855 [41:59<1:20:46,  2.76s/it, loss=0.0841, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1102/2855 [41:59<1:16:00,  2.60s/it, loss=0.0841, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1102/2855 [42:01<1:16:00,  2.60s/it, loss=0.1411, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1103/2855 [42:01<1:12:13,  2.47s/it, loss=0.1411, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1103/2855 [42:03<1:12:13,  2.47s/it, loss=0.1590, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1104/2855 [42:03<1:10:35,  2.42s/it, loss=0.1590, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1104/2855 [42:05<1:10:35,  2.42s/it, loss=0.1174, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1105/2855 [42:05<1:08:46,  2.36s/it, loss=0.1174, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1105/2855 [42:08<1:08:46,  2.36s/it, loss=0.2591, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1106/2855 [42:08<1:07:53,  2.33s/it, loss=0.2591, lr=2.79e-05]
Epoch 2/8:  39%|███▊      | 1106/2855 [42:10<1:07:53,  2.33s/it, loss=0.1513, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1107/2855 [42:10<1:06:10,  2.27s/it, loss=0.1513, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1107/2855 [42:12<1:06:10,  2.27s/it, loss=0.1993, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1108/2855 [42:12<1:05:11,  2.24s/it, loss=0.1993, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1108/2855 [42:14<1:05:11,  2.24s/it, loss=0.1616, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1109/2855 [42:14<1:04:42,  2.22s/it, loss=0.1616, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1109/2855 [42:16<1:04:42,  2.22s/it, loss=0.1640, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1110/2855 [42:16<1:04:00,  2.20s/it, loss=0.1640, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1110/2855 [42:19<1:04:00,  2.20s/it, loss=0.2293, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1111/2855 [42:19<1:06:29,  2.29s/it, loss=0.2293, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1111/2855 [42:21<1:06:29,  2.29s/it, loss=0.2142, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1112/2855 [42:21<1:05:22,  2.25s/it, loss=0.2142, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1112/2855 [42:23<1:05:22,  2.25s/it, loss=0.2405, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1113/2855 [42:23<1:05:28,  2.26s/it, loss=0.2405, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1113/2855 [42:26<1:05:28,  2.26s/it, loss=0.2231, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1114/2855 [42:26<1:06:22,  2.29s/it, loss=0.2231, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1114/2855 [42:28<1:06:22,  2.29s/it, loss=0.0819, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1115/2855 [42:28<1:07:48,  2.34s/it, loss=0.0819, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1115/2855 [42:30<1:07:48,  2.34s/it, loss=0.1122, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1116/2855 [42:30<1:07:11,  2.32s/it, loss=0.1122, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1116/2855 [42:33<1:07:11,  2.32s/it, loss=0.1790, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1117/2855 [42:33<1:08:06,  2.35s/it, loss=0.1790, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1117/2855 [42:35<1:08:06,  2.35s/it, loss=0.0847, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1118/2855 [42:35<1:08:00,  2.35s/it, loss=0.0847, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1118/2855 [42:37<1:08:00,  2.35s/it, loss=0.1367, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1119/2855 [42:37<1:06:58,  2.31s/it, loss=0.1367, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1119/2855 [42:40<1:06:58,  2.31s/it, loss=0.2341, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1120/2855 [42:40<1:08:06,  2.36s/it, loss=0.2341, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1120/2855 [42:42<1:08:06,  2.36s/it, loss=0.1199, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1121/2855 [42:42<1:07:53,  2.35s/it, loss=0.1199, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1121/2855 [42:45<1:07:53,  2.35s/it, loss=0.2046, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1122/2855 [42:45<1:09:17,  2.40s/it, loss=0.2046, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1122/2855 [42:47<1:09:17,  2.40s/it, loss=0.1635, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1123/2855 [42:47<1:07:22,  2.33s/it, loss=0.1635, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1123/2855 [42:49<1:07:22,  2.33s/it, loss=0.1166, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1124/2855 [42:49<1:08:24,  2.37s/it, loss=0.1166, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1124/2855 [42:51<1:08:24,  2.37s/it, loss=0.1716, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1125/2855 [42:51<1:06:49,  2.32s/it, loss=0.1716, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1125/2855 [42:54<1:06:49,  2.32s/it, loss=0.0522, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1126/2855 [42:54<1:08:30,  2.38s/it, loss=0.0522, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1126/2855 [42:56<1:08:30,  2.38s/it, loss=0.1522, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1127/2855 [42:56<1:07:50,  2.36s/it, loss=0.1522, lr=2.79e-05]
Epoch 2/8:  39%|███▉      | 1127/2855 [42:59<1:07:50,  2.36s/it, loss=0.2838, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1128/2855 [42:59<1:09:44,  2.42s/it, loss=0.2838, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1128/2855 [43:01<1:09:44,  2.42s/it, loss=0.1683, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1129/2855 [43:01<1:09:30,  2.42s/it, loss=0.1683, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1129/2855 [43:04<1:09:30,  2.42s/it, loss=0.1835, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1130/2855 [43:04<1:08:40,  2.39s/it, loss=0.1835, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1130/2855 [43:06<1:08:40,  2.39s/it, loss=0.1613, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1131/2855 [43:06<1:08:28,  2.38s/it, loss=0.1613, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1131/2855 [43:08<1:08:28,  2.38s/it, loss=0.0719, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1132/2855 [43:08<1:09:11,  2.41s/it, loss=0.0719, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1132/2855 [43:11<1:09:11,  2.41s/it, loss=0.3820, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1133/2855 [43:11<1:09:08,  2.41s/it, loss=0.3820, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1133/2855 [43:14<1:09:08,  2.41s/it, loss=0.2758, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1134/2855 [43:14<1:11:50,  2.50s/it, loss=0.2758, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1134/2855 [43:16<1:11:50,  2.50s/it, loss=0.2923, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1135/2855 [43:16<1:09:36,  2.43s/it, loss=0.2923, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1135/2855 [43:18<1:09:36,  2.43s/it, loss=0.1662, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1136/2855 [43:18<1:09:25,  2.42s/it, loss=0.1662, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1136/2855 [43:21<1:09:25,  2.42s/it, loss=0.2792, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1137/2855 [43:21<1:15:02,  2.62s/it, loss=0.2792, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1137/2855 [43:23<1:15:02,  2.62s/it, loss=0.1489, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1138/2855 [43:23<1:11:00,  2.48s/it, loss=0.1489, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1138/2855 [43:26<1:11:00,  2.48s/it, loss=0.0890, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1139/2855 [43:26<1:10:04,  2.45s/it, loss=0.0890, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1139/2855 [43:28<1:10:04,  2.45s/it, loss=0.0411, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1140/2855 [43:28<1:09:20,  2.43s/it, loss=0.0411, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1140/2855 [43:31<1:09:20,  2.43s/it, loss=0.1500, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1141/2855 [43:31<1:08:41,  2.40s/it, loss=0.1500, lr=2.79e-05]
Epoch 2/8:  40%|███▉      | 1141/2855 [43:33<1:08:41,  2.40s/it, loss=0.1204, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1142/2855 [43:33<1:07:45,  2.37s/it, loss=0.1204, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1142/2855 [43:35<1:07:45,  2.37s/it, loss=0.0699, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1143/2855 [43:35<1:07:39,  2.37s/it, loss=0.0699, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1143/2855 [43:37<1:07:39,  2.37s/it, loss=0.1020, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1144/2855 [43:37<1:06:10,  2.32s/it, loss=0.1020, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1144/2855 [43:40<1:06:10,  2.32s/it, loss=0.0112, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1145/2855 [43:40<1:06:38,  2.34s/it, loss=0.0112, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1145/2855 [43:42<1:06:38,  2.34s/it, loss=0.1709, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1146/2855 [43:42<1:06:10,  2.32s/it, loss=0.1709, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1146/2855 [43:45<1:06:10,  2.32s/it, loss=0.2077, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1147/2855 [43:45<1:07:04,  2.36s/it, loss=0.2077, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1147/2855 [43:47<1:07:04,  2.36s/it, loss=0.1163, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1148/2855 [43:47<1:06:51,  2.35s/it, loss=0.1163, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1148/2855 [43:49<1:06:51,  2.35s/it, loss=0.1764, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1149/2855 [43:49<1:07:51,  2.39s/it, loss=0.1764, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1149/2855 [43:52<1:07:51,  2.39s/it, loss=0.2307, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1150/2855 [43:52<1:06:44,  2.35s/it, loss=0.2307, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1150/2855 [43:54<1:06:44,  2.35s/it, loss=0.1964, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1151/2855 [43:54<1:07:50,  2.39s/it, loss=0.1964, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1151/2855 [43:56<1:07:50,  2.39s/it, loss=0.2186, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1152/2855 [43:56<1:06:23,  2.34s/it, loss=0.2186, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1152/2855 [43:59<1:06:23,  2.34s/it, loss=0.1966, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1153/2855 [43:59<1:06:15,  2.34s/it, loss=0.1966, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1153/2855 [44:01<1:06:15,  2.34s/it, loss=0.1677, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1154/2855 [44:01<1:06:19,  2.34s/it, loss=0.1677, lr=2.79e-05]
Epoch 2/8:  40%|████      | 1154/2855 [44:03<1:06:19,  2.34s/it, loss=0.1693, lr=2.78e-05]
Epoch 2/8:  40%|████      | 1155/2855 [44:03<1:05:28,  2.31s/it, loss=0.1693, lr=2.78e-05]
Epoch 2/8:  40%|████      | 1155/2855 [44:06<1:05:28,  2.31s/it, loss=0.1837, lr=2.78e-05]
Epoch 2/8:  40%|████      | 1156/2855 [44:06<1:09:25,  2.45s/it, loss=0.1837, lr=2.78e-05]
Epoch 2/8:  40%|████      | 1156/2855 [44:08<1:09:25,  2.45s/it, loss=0.1952, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1157/2855 [44:08<1:07:28,  2.38s/it, loss=0.1952, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1157/2855 [44:11<1:07:28,  2.38s/it, loss=0.1858, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1158/2855 [44:11<1:07:55,  2.40s/it, loss=0.1858, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1158/2855 [44:13<1:07:55,  2.40s/it, loss=0.2041, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1159/2855 [44:13<1:06:13,  2.34s/it, loss=0.2041, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1159/2855 [44:15<1:06:13,  2.34s/it, loss=0.2845, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1160/2855 [44:15<1:06:19,  2.35s/it, loss=0.2845, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1160/2855 [44:18<1:06:19,  2.35s/it, loss=0.1692, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1161/2855 [44:18<1:05:53,  2.33s/it, loss=0.1692, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1161/2855 [44:20<1:05:53,  2.33s/it, loss=0.0838, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1162/2855 [44:20<1:06:22,  2.35s/it, loss=0.0838, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1162/2855 [44:22<1:06:22,  2.35s/it, loss=0.0735, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1163/2855 [44:22<1:05:58,  2.34s/it, loss=0.0735, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1163/2855 [44:25<1:05:58,  2.34s/it, loss=0.2133, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1164/2855 [44:25<1:07:28,  2.39s/it, loss=0.2133, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1164/2855 [44:27<1:07:28,  2.39s/it, loss=0.1271, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1165/2855 [44:27<1:06:22,  2.36s/it, loss=0.1271, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1165/2855 [44:30<1:06:22,  2.36s/it, loss=0.1590, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1166/2855 [44:30<1:07:08,  2.38s/it, loss=0.1590, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1166/2855 [44:32<1:07:08,  2.38s/it, loss=0.2759, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1167/2855 [44:32<1:07:00,  2.38s/it, loss=0.2759, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1167/2855 [44:34<1:07:00,  2.38s/it, loss=0.1555, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1168/2855 [44:34<1:06:51,  2.38s/it, loss=0.1555, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1168/2855 [44:37<1:06:51,  2.38s/it, loss=0.0699, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1169/2855 [44:37<1:06:20,  2.36s/it, loss=0.0699, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1169/2855 [44:39<1:06:20,  2.36s/it, loss=0.0787, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1170/2855 [44:39<1:05:16,  2.32s/it, loss=0.0787, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1170/2855 [44:41<1:05:16,  2.32s/it, loss=0.2533, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1171/2855 [44:41<1:05:28,  2.33s/it, loss=0.2533, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1171/2855 [44:43<1:05:28,  2.33s/it, loss=0.2809, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1172/2855 [44:43<1:04:30,  2.30s/it, loss=0.2809, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1172/2855 [44:46<1:04:30,  2.30s/it, loss=0.1934, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1173/2855 [44:46<1:05:29,  2.34s/it, loss=0.1934, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1173/2855 [44:48<1:05:29,  2.34s/it, loss=0.2211, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1174/2855 [44:48<1:05:56,  2.35s/it, loss=0.2211, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1174/2855 [44:51<1:05:56,  2.35s/it, loss=0.1234, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1175/2855 [44:51<1:06:27,  2.37s/it, loss=0.1234, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1175/2855 [44:53<1:06:27,  2.37s/it, loss=0.3390, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1176/2855 [44:53<1:03:49,  2.28s/it, loss=0.3390, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1176/2855 [44:55<1:03:49,  2.28s/it, loss=0.1830, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1177/2855 [44:55<1:06:51,  2.39s/it, loss=0.1830, lr=2.78e-05]
Epoch 2/8:  41%|████      | 1177/2855 [44:58<1:06:51,  2.39s/it, loss=0.1094, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1178/2855 [44:58<1:06:40,  2.39s/it, loss=0.1094, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1178/2855 [45:00<1:06:40,  2.39s/it, loss=0.1793, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1179/2855 [45:00<1:06:58,  2.40s/it, loss=0.1793, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1179/2855 [45:02<1:06:58,  2.40s/it, loss=0.1391, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1180/2855 [45:02<1:05:57,  2.36s/it, loss=0.1391, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1180/2855 [45:05<1:05:57,  2.36s/it, loss=0.1513, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1181/2855 [45:05<1:07:45,  2.43s/it, loss=0.1513, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1181/2855 [45:07<1:07:45,  2.43s/it, loss=0.0904, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1182/2855 [45:07<1:06:32,  2.39s/it, loss=0.0904, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1182/2855 [45:10<1:06:32,  2.39s/it, loss=0.1216, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1183/2855 [45:10<1:09:45,  2.50s/it, loss=0.1216, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1183/2855 [45:12<1:09:45,  2.50s/it, loss=0.0354, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1184/2855 [45:12<1:07:29,  2.42s/it, loss=0.0354, lr=2.78e-05]
Epoch 2/8:  41%|████▏     | 1184/2855 [45:15<1:07:29,  2.42s/it, loss=0.1885, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1185/2855 [45:15<1:07:49,  2.44s/it, loss=0.1885, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1185/2855 [45:17<1:07:49,  2.44s/it, loss=0.1354, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1186/2855 [45:17<1:05:33,  2.36s/it, loss=0.1354, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1186/2855 [45:19<1:05:33,  2.36s/it, loss=0.1956, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1187/2855 [45:19<1:03:15,  2.28s/it, loss=0.1956, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1187/2855 [45:21<1:03:15,  2.28s/it, loss=0.2156, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1188/2855 [45:21<1:02:05,  2.23s/it, loss=0.2156, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1188/2855 [45:23<1:02:05,  2.23s/it, loss=0.0654, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1189/2855 [45:23<1:02:16,  2.24s/it, loss=0.0654, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1189/2855 [45:26<1:02:16,  2.24s/it, loss=0.0993, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1190/2855 [45:26<1:02:27,  2.25s/it, loss=0.0993, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1190/2855 [45:28<1:02:27,  2.25s/it, loss=0.1149, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1191/2855 [45:28<1:01:35,  2.22s/it, loss=0.1149, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1191/2855 [45:30<1:01:35,  2.22s/it, loss=0.1745, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1192/2855 [45:30<1:02:52,  2.27s/it, loss=0.1745, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1192/2855 [45:32<1:02:52,  2.27s/it, loss=0.2078, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1193/2855 [45:32<1:02:27,  2.25s/it, loss=0.2078, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1193/2855 [45:35<1:02:27,  2.25s/it, loss=0.0873, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1194/2855 [45:35<1:04:47,  2.34s/it, loss=0.0873, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1194/2855 [45:37<1:04:47,  2.34s/it, loss=0.1281, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1195/2855 [45:37<1:03:35,  2.30s/it, loss=0.1281, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1195/2855 [45:40<1:03:35,  2.30s/it, loss=0.0670, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1196/2855 [45:40<1:05:11,  2.36s/it, loss=0.0670, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1196/2855 [45:42<1:05:11,  2.36s/it, loss=0.2676, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1197/2855 [45:42<1:04:00,  2.32s/it, loss=0.2676, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1197/2855 [45:44<1:04:00,  2.32s/it, loss=0.1877, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1198/2855 [45:44<1:05:02,  2.36s/it, loss=0.1877, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1198/2855 [45:47<1:05:02,  2.36s/it, loss=0.1481, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1199/2855 [45:47<1:04:05,  2.32s/it, loss=0.1481, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1199/2855 [45:49<1:04:05,  2.32s/it, loss=0.1797, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1200/2855 [45:49<1:03:37,  2.31s/it, loss=0.1797, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1200/2855 [45:51<1:03:37,  2.31s/it, loss=0.0612, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1201/2855 [45:51<1:05:30,  2.38s/it, loss=0.0612, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1201/2855 [45:54<1:05:30,  2.38s/it, loss=0.1520, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1202/2855 [45:54<1:04:29,  2.34s/it, loss=0.1520, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1202/2855 [45:56<1:04:29,  2.34s/it, loss=0.1010, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1203/2855 [45:56<1:05:37,  2.38s/it, loss=0.1010, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1203/2855 [45:58<1:05:37,  2.38s/it, loss=0.1895, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1204/2855 [45:58<1:05:08,  2.37s/it, loss=0.1895, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1204/2855 [46:01<1:05:08,  2.37s/it, loss=0.1630, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1205/2855 [46:01<1:06:49,  2.43s/it, loss=0.1630, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1205/2855 [46:03<1:06:49,  2.43s/it, loss=0.1121, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1206/2855 [46:03<1:04:44,  2.36s/it, loss=0.1121, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1206/2855 [46:06<1:04:44,  2.36s/it, loss=0.1018, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1207/2855 [46:06<1:06:42,  2.43s/it, loss=0.1018, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1207/2855 [46:08<1:06:42,  2.43s/it, loss=0.2051, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1208/2855 [46:08<1:06:28,  2.42s/it, loss=0.2051, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1208/2855 [46:11<1:06:28,  2.42s/it, loss=0.1768, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1209/2855 [46:11<1:07:31,  2.46s/it, loss=0.1768, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1209/2855 [46:13<1:07:31,  2.46s/it, loss=0.0754, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1210/2855 [46:13<1:05:53,  2.40s/it, loss=0.0754, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1210/2855 [46:16<1:05:53,  2.40s/it, loss=0.1340, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1211/2855 [46:16<1:09:22,  2.53s/it, loss=0.1340, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1211/2855 [46:18<1:09:22,  2.53s/it, loss=0.0771, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1212/2855 [46:18<1:06:37,  2.43s/it, loss=0.0771, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1212/2855 [46:21<1:06:37,  2.43s/it, loss=0.1614, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1213/2855 [46:21<1:07:52,  2.48s/it, loss=0.1614, lr=2.78e-05]
Epoch 2/8:  42%|████▏     | 1213/2855 [46:23<1:07:52,  2.48s/it, loss=0.2172, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1214/2855 [46:23<1:06:47,  2.44s/it, loss=0.2172, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1214/2855 [46:26<1:06:47,  2.44s/it, loss=0.0041, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1215/2855 [46:26<1:10:12,  2.57s/it, loss=0.0041, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1215/2855 [46:33<1:10:12,  2.57s/it, loss=0.2440, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1216/2855 [46:33<1:47:29,  3.94s/it, loss=0.2440, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1216/2855 [46:36<1:47:29,  3.94s/it, loss=0.2018, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1217/2855 [46:36<1:41:46,  3.73s/it, loss=0.2018, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1217/2855 [46:39<1:41:46,  3.73s/it, loss=0.0428, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1218/2855 [46:39<1:30:31,  3.32s/it, loss=0.0428, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1218/2855 [46:41<1:30:31,  3.32s/it, loss=0.1535, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1219/2855 [46:41<1:23:15,  3.05s/it, loss=0.1535, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1219/2855 [46:43<1:23:15,  3.05s/it, loss=0.0868, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1220/2855 [46:43<1:15:49,  2.78s/it, loss=0.0868, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1220/2855 [46:46<1:15:49,  2.78s/it, loss=0.2141, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1221/2855 [46:46<1:14:02,  2.72s/it, loss=0.2141, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1221/2855 [46:48<1:14:02,  2.72s/it, loss=0.1362, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1222/2855 [46:48<1:08:21,  2.51s/it, loss=0.1362, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1222/2855 [46:50<1:08:21,  2.51s/it, loss=0.2261, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1223/2855 [46:50<1:08:21,  2.51s/it, loss=0.2261, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1223/2855 [46:53<1:08:21,  2.51s/it, loss=0.1263, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1224/2855 [46:53<1:05:30,  2.41s/it, loss=0.1263, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1224/2855 [46:55<1:05:30,  2.41s/it, loss=0.1353, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1225/2855 [46:55<1:06:18,  2.44s/it, loss=0.1353, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1225/2855 [46:57<1:06:18,  2.44s/it, loss=0.1844, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1226/2855 [46:57<1:03:35,  2.34s/it, loss=0.1844, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1226/2855 [46:59<1:03:35,  2.34s/it, loss=0.1960, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1227/2855 [46:59<1:01:42,  2.27s/it, loss=0.1960, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1227/2855 [47:02<1:01:42,  2.27s/it, loss=0.0876, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1228/2855 [47:02<1:03:21,  2.34s/it, loss=0.0876, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1228/2855 [47:04<1:03:21,  2.34s/it, loss=0.0816, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1229/2855 [47:04<1:01:42,  2.28s/it, loss=0.0816, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1229/2855 [47:06<1:01:42,  2.28s/it, loss=0.1230, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1230/2855 [47:06<1:03:37,  2.35s/it, loss=0.1230, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1230/2855 [47:08<1:03:37,  2.35s/it, loss=0.0459, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1231/2855 [47:08<1:01:06,  2.26s/it, loss=0.0459, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1231/2855 [47:11<1:01:06,  2.26s/it, loss=0.0694, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1232/2855 [47:11<1:02:06,  2.30s/it, loss=0.0694, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1232/2855 [47:13<1:02:06,  2.30s/it, loss=0.1471, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1233/2855 [47:13<58:43,  2.17s/it, loss=0.1471, lr=2.78e-05]  
Epoch 2/8:  43%|████▎     | 1233/2855 [47:15<58:43,  2.17s/it, loss=0.1353, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1234/2855 [47:15<59:53,  2.22s/it, loss=0.1353, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1234/2855 [47:17<59:53,  2.22s/it, loss=0.1446, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1235/2855 [47:17<1:01:11,  2.27s/it, loss=0.1446, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1235/2855 [47:20<1:01:11,  2.27s/it, loss=0.1763, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1236/2855 [47:20<1:04:53,  2.40s/it, loss=0.1763, lr=2.78e-05]
Epoch 2/8:  43%|██���█▎     | 1236/2855 [47:22<1:04:53,  2.40s/it, loss=0.0802, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1237/2855 [47:22<1:03:15,  2.35s/it, loss=0.0802, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1237/2855 [47:25<1:03:15,  2.35s/it, loss=0.1558, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1238/2855 [47:25<1:03:09,  2.34s/it, loss=0.1558, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1238/2855 [47:27<1:03:09,  2.34s/it, loss=0.2538, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1239/2855 [47:27<1:04:08,  2.38s/it, loss=0.2538, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1239/2855 [47:29<1:04:08,  2.38s/it, loss=0.1814, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1240/2855 [47:29<1:02:42,  2.33s/it, loss=0.1814, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1240/2855 [47:32<1:02:42,  2.33s/it, loss=0.2337, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1241/2855 [47:32<1:02:33,  2.33s/it, loss=0.2337, lr=2.78e-05]
Epoch 2/8:  43%|████▎     | 1241/2855 [47:34<1:02:33,  2.33s/it, loss=0.0783, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1242/2855 [47:34<1:00:40,  2.26s/it, loss=0.0783, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1242/2855 [47:36<1:00:40,  2.26s/it, loss=0.2231, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1243/2855 [47:36<1:00:58,  2.27s/it, loss=0.2231, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1243/2855 [47:38<1:00:58,  2.27s/it, loss=0.2012, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1244/2855 [47:38<59:31,  2.22s/it, loss=0.2012, lr=2.78e-05]  
Epoch 2/8:  44%|████▎     | 1244/2855 [47:40<59:31,  2.22s/it, loss=0.0932, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1245/2855 [47:40<59:13,  2.21s/it, loss=0.0932, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1245/2855 [47:42<59:13,  2.21s/it, loss=0.0802, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1246/2855 [47:42<57:29,  2.14s/it, loss=0.0802, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1246/2855 [47:44<57:29,  2.14s/it, loss=0.1216, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1247/2855 [47:44<55:03,  2.05s/it, loss=0.1216, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1247/2855 [47:46<55:03,  2.05s/it, loss=0.0405, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1248/2855 [47:46<56:03,  2.09s/it, loss=0.0405, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1248/2855 [47:48<56:03,  2.09s/it, loss=0.1256, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1249/2855 [47:48<55:44,  2.08s/it, loss=0.1256, lr=2.78e-05]
Epoch 2/8:  44%|████▎     | 1249/2855 [47:51<55:44,  2.08s/it, loss=0.2305, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1250/2855 [47:51<57:25,  2.15s/it, loss=0.2305, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1250/2855 [47:53<57:25,  2.15s/it, loss=0.1533, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1251/2855 [47:53<55:34,  2.08s/it, loss=0.1533, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1251/2855 [47:54<55:34,  2.08s/it, loss=0.1721, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1252/2855 [47:54<52:28,  1.96s/it, loss=0.1721, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1252/2855 [47:56<52:28,  1.96s/it, loss=0.0538, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1253/2855 [47:56<53:03,  1.99s/it, loss=0.0538, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1253/2855 [47:58<53:03,  1.99s/it, loss=0.2280, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1254/2855 [47:58<52:57,  1.98s/it, loss=0.2280, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1254/2855 [48:00<52:57,  1.98s/it, loss=0.1628, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1255/2855 [48:00<53:47,  2.02s/it, loss=0.1628, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1255/2855 [48:05<53:47,  2.02s/it, loss=0.0416, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1256/2855 [48:05<1:13:08,  2.74s/it, loss=0.0416, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1256/2855 [48:07<1:13:08,  2.74s/it, loss=0.1433, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1257/2855 [48:07<1:05:50,  2.47s/it, loss=0.1433, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1257/2855 [48:09<1:05:50,  2.47s/it, loss=0.1654, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1258/2855 [48:09<1:03:26,  2.38s/it, loss=0.1654, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1258/2855 [48:11<1:03:26,  2.38s/it, loss=0.1826, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1259/2855 [48:11<59:16,  2.23s/it, loss=0.1826, lr=2.77e-05]  
Epoch 2/8:  44%|████▍     | 1259/2855 [48:13<59:16,  2.23s/it, loss=0.2061, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1260/2855 [48:13<56:08,  2.11s/it, loss=0.2061, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1260/2855 [48:14<56:08,  2.11s/it, loss=0.1834, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1261/2855 [48:14<53:51,  2.03s/it, loss=0.1834, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1261/2855 [48:16<53:51,  2.03s/it, loss=0.1392, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1262/2855 [48:16<52:31,  1.98s/it, loss=0.1392, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1262/2855 [48:18<52:31,  1.98s/it, loss=0.2172, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1263/2855 [48:18<51:47,  1.95s/it, loss=0.2172, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1263/2855 [48:20<51:47,  1.95s/it, loss=0.1942, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1264/2855 [48:20<50:36,  1.91s/it, loss=0.1942, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1264/2855 [48:22<50:36,  1.91s/it, loss=0.0331, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1265/2855 [48:22<48:21,  1.83s/it, loss=0.0331, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1265/2855 [48:24<48:21,  1.83s/it, loss=0.0881, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1266/2855 [48:24<49:19,  1.86s/it, loss=0.0881, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1266/2855 [48:25<49:19,  1.86s/it, loss=0.1406, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1267/2855 [48:25<48:33,  1.83s/it, loss=0.1406, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1267/2855 [48:27<48:33,  1.83s/it, loss=0.0958, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1268/2855 [48:27<46:28,  1.76s/it, loss=0.0958, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1268/2855 [48:29<46:28,  1.76s/it, loss=0.0615, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1269/2855 [48:29<45:16,  1.71s/it, loss=0.0615, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1269/2855 [48:30<45:16,  1.71s/it, loss=0.2513, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1270/2855 [48:30<44:51,  1.70s/it, loss=0.2513, lr=2.77e-05]
Epoch 2/8:  44%|████▍     | 1270/2855 [48:32<44:51,  1.70s/it, loss=0.0661, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1271/2855 [48:32<46:06,  1.75s/it, loss=0.0661, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1271/2855 [48:34<46:06,  1.75s/it, loss=0.1234, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1272/2855 [48:34<47:05,  1.79s/it, loss=0.1234, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1272/2855 [48:36<47:05,  1.79s/it, loss=0.0811, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1273/2855 [48:36<47:59,  1.82s/it, loss=0.0811, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1273/2855 [48:38<47:59,  1.82s/it, loss=0.3466, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1274/2855 [48:38<47:19,  1.80s/it, loss=0.3466, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1274/2855 [48:39<47:19,  1.80s/it, loss=0.1510, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1275/2855 [48:39<46:42,  1.77s/it, loss=0.1510, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1275/2855 [48:41<46:42,  1.77s/it, loss=0.1308, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1276/2855 [48:41<47:40,  1.81s/it, loss=0.1308, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1276/2855 [48:43<47:40,  1.81s/it, loss=0.1529, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1277/2855 [48:43<47:07,  1.79s/it, loss=0.1529, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1277/2855 [48:45<47:07,  1.79s/it, loss=0.1591, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1278/2855 [48:45<46:05,  1.75s/it, loss=0.1591, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1278/2855 [48:46<46:05,  1.75s/it, loss=0.1237, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1279/2855 [48:46<46:20,  1.76s/it, loss=0.1237, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1279/2855 [48:48<46:20,  1.76s/it, loss=0.1754, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1280/2855 [48:48<46:25,  1.77s/it, loss=0.1754, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1280/2855 [48:50<46:25,  1.77s/it, loss=0.2764, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1281/2855 [48:50<45:51,  1.75s/it, loss=0.2764, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1281/2855 [48:52<45:51,  1.75s/it, loss=0.3004, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1282/2855 [48:52<45:38,  1.74s/it, loss=0.3004, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1282/2855 [48:53<45:38,  1.74s/it, loss=0.0067, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1283/2855 [48:53<45:29,  1.74s/it, loss=0.0067, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1283/2855 [48:55<45:29,  1.74s/it, loss=0.1250, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1284/2855 [48:55<44:24,  1.70s/it, loss=0.1250, lr=2.77e-05]
Epoch 2/8:  45%|████▍     | 1284/2855 [48:56<44:24,  1.70s/it, loss=0.2070, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1285/2855 [48:56<43:13,  1.65s/it, loss=0.2070, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1285/2855 [48:58<43:13,  1.65s/it, loss=0.1444, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1286/2855 [48:58<43:25,  1.66s/it, loss=0.1444, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1286/2855 [49:00<43:25,  1.66s/it, loss=0.1543, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1287/2855 [49:00<46:35,  1.78s/it, loss=0.1543, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1287/2855 [49:03<46:35,  1.78s/it, loss=0.1568, lr=2.77e-05]
Epoch 2/8:  45%|█���██▌     | 1288/2855 [49:03<51:02,  1.95s/it, loss=0.1568, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1288/2855 [49:05<51:02,  1.95s/it, loss=0.2019, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1289/2855 [49:05<55:22,  2.12s/it, loss=0.2019, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1289/2855 [49:07<55:22,  2.12s/it, loss=0.1725, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1290/2855 [49:07<55:33,  2.13s/it, loss=0.1725, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1290/2855 [49:09<55:33,  2.13s/it, loss=0.0310, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1291/2855 [49:09<56:00,  2.15s/it, loss=0.0310, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1291/2855 [49:12<56:00,  2.15s/it, loss=0.1789, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1292/2855 [49:12<56:57,  2.19s/it, loss=0.1789, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1292/2855 [49:14<56:57,  2.19s/it, loss=0.1005, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1293/2855 [49:14<54:32,  2.10s/it, loss=0.1005, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1293/2855 [49:15<54:32,  2.10s/it, loss=0.1160, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1294/2855 [49:15<52:22,  2.01s/it, loss=0.1160, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1294/2855 [49:17<52:22,  2.01s/it, loss=0.1789, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1295/2855 [49:17<50:36,  1.95s/it, loss=0.1789, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1295/2855 [49:20<50:36,  1.95s/it, loss=0.1944, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1296/2855 [49:20<53:33,  2.06s/it, loss=0.1944, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1296/2855 [49:21<53:33,  2.06s/it, loss=0.2924, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1297/2855 [49:21<52:27,  2.02s/it, loss=0.2924, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1297/2855 [49:23<52:27,  2.02s/it, loss=0.2164, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1298/2855 [49:23<49:43,  1.92s/it, loss=0.2164, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1298/2855 [49:25<49:43,  1.92s/it, loss=0.3421, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1299/2855 [49:25<47:10,  1.82s/it, loss=0.3421, lr=2.77e-05]
Epoch 2/8:  45%|████▌     | 1299/2855 [49:27<47:10,  1.82s/it, loss=0.0849, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1300/2855 [49:27<49:21,  1.90s/it, loss=0.0849, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1300/2855 [49:29<49:21,  1.90s/it, loss=0.2189, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1301/2855 [49:29<47:26,  1.83s/it, loss=0.2189, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1301/2855 [49:30<47:26,  1.83s/it, loss=0.1686, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1302/2855 [49:30<46:37,  1.80s/it, loss=0.1686, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1302/2855 [49:32<46:37,  1.80s/it, loss=0.0785, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1303/2855 [49:32<46:41,  1.81s/it, loss=0.0785, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1303/2855 [49:34<46:41,  1.81s/it, loss=0.1325, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1304/2855 [49:34<44:48,  1.73s/it, loss=0.1325, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1304/2855 [49:35<44:48,  1.73s/it, loss=0.0928, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1305/2855 [49:35<44:42,  1.73s/it, loss=0.0928, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1305/2855 [49:37<44:42,  1.73s/it, loss=0.1194, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1306/2855 [49:37<45:29,  1.76s/it, loss=0.1194, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1306/2855 [49:39<45:29,  1.76s/it, loss=0.0439, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1307/2855 [49:39<44:07,  1.71s/it, loss=0.0439, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1307/2855 [49:40<44:07,  1.71s/it, loss=0.3109, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1308/2855 [49:40<43:15,  1.68s/it, loss=0.3109, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1308/2855 [49:42<43:15,  1.68s/it, loss=0.1940, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1309/2855 [49:42<43:31,  1.69s/it, loss=0.1940, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1309/2855 [49:44<43:31,  1.69s/it, loss=0.0597, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1310/2855 [49:44<41:59,  1.63s/it, loss=0.0597, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1310/2855 [49:45<41:59,  1.63s/it, loss=0.1441, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1311/2855 [49:45<40:44,  1.58s/it, loss=0.1441, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1311/2855 [49:47<40:44,  1.58s/it, loss=0.1580, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1312/2855 [49:47<42:52,  1.67s/it, loss=0.1580, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1312/2855 [49:49<42:52,  1.67s/it, loss=0.1227, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1313/2855 [49:49<43:19,  1.69s/it, loss=0.1227, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1313/2855 [49:50<43:19,  1.69s/it, loss=0.2436, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1314/2855 [49:50<43:15,  1.68s/it, loss=0.2436, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1314/2855 [49:52<43:15,  1.68s/it, loss=0.1140, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1315/2855 [49:52<43:24,  1.69s/it, loss=0.1140, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1315/2855 [49:53<43:24,  1.69s/it, loss=0.0544, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1316/2855 [49:53<41:35,  1.62s/it, loss=0.0544, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1316/2855 [49:55<41:35,  1.62s/it, loss=0.1046, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1317/2855 [49:55<40:28,  1.58s/it, loss=0.1046, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1317/2855 [49:57<40:28,  1.58s/it, loss=0.1870, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1318/2855 [49:57<43:27,  1.70s/it, loss=0.1870, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1318/2855 [49:59<43:27,  1.70s/it, loss=0.1035, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1319/2855 [49:59<43:03,  1.68s/it, loss=0.1035, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1319/2855 [50:00<43:03,  1.68s/it, loss=0.2121, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1320/2855 [50:00<43:42,  1.71s/it, loss=0.2121, lr=2.77e-05]
Epoch 2/8:  46%|████▌     | 1320/2855 [50:02<43:42,  1.71s/it, loss=0.2196, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1321/2855 [50:02<45:16,  1.77s/it, loss=0.2196, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1321/2855 [50:04<45:16,  1.77s/it, loss=0.1591, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1322/2855 [50:04<45:03,  1.76s/it, loss=0.1591, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1322/2855 [50:06<45:03,  1.76s/it, loss=0.1478, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1323/2855 [50:06<45:56,  1.80s/it, loss=0.1478, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1323/2855 [50:08<45:56,  1.80s/it, loss=0.0924, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1324/2855 [50:08<44:49,  1.76s/it, loss=0.0924, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1324/2855 [50:09<44:49,  1.76s/it, loss=0.3275, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1325/2855 [50:09<43:03,  1.69s/it, loss=0.3275, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1325/2855 [50:11<43:03,  1.69s/it, loss=0.1254, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1326/2855 [50:11<45:31,  1.79s/it, loss=0.1254, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1326/2855 [50:13<45:31,  1.79s/it, loss=0.1594, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1327/2855 [50:13<46:04,  1.81s/it, loss=0.1594, lr=2.77e-05]
Epoch 2/8:  46%|████▋     | 1327/2855 [50:15<46:04,  1.81s/it, loss=0.1372, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1328/2855 [50:15<45:37,  1.79s/it, loss=0.1372, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1328/2855 [50:17<45:37,  1.79s/it, loss=0.1499, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1329/2855 [50:17<46:22,  1.82s/it, loss=0.1499, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1329/2855 [50:19<46:22,  1.82s/it, loss=0.0001, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1330/2855 [50:19<47:35,  1.87s/it, loss=0.0001, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1330/2855 [50:20<47:35,  1.87s/it, loss=0.0395, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1331/2855 [50:20<46:48,  1.84s/it, loss=0.0395, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1331/2855 [50:22<46:48,  1.84s/it, loss=0.1806, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1332/2855 [50:22<46:45,  1.84s/it, loss=0.1806, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1332/2855 [50:24<46:45,  1.84s/it, loss=0.1095, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1333/2855 [50:24<44:44,  1.76s/it, loss=0.1095, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1333/2855 [50:25<44:44,  1.76s/it, loss=0.1122, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1334/2855 [50:25<42:43,  1.69s/it, loss=0.1122, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1334/2855 [50:27<42:43,  1.69s/it, loss=0.0563, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1335/2855 [50:27<46:03,  1.82s/it, loss=0.0563, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1335/2855 [50:29<46:03,  1.82s/it, loss=0.1429, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1336/2855 [50:29<42:27,  1.68s/it, loss=0.1429, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1336/2855 [50:30<42:27,  1.68s/it, loss=0.1713, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1337/2855 [50:30<40:48,  1.61s/it, loss=0.1713, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1337/2855 [50:32<40:48,  1.61s/it, loss=0.1669, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1338/2855 [50:32<44:02,  1.74s/it, loss=0.1669, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1338/2855 [50:34<44:02,  1.74s/it, loss=0.1397, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1339/2855 [50:34<43:49,  1.73s/it, loss=0.1397, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1339/2855 [50:36<43:49,  1.73s/it, loss=0.2406, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1340/2855 [50:36<45:27,  1.80s/it, loss=0.2406, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1340/2855 [50:38<45:27,  1.80s/it, loss=0.1413, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1341/2855 [50:38<44:34,  1.77s/it, loss=0.1413, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1341/2855 [50:39<44:34,  1.77s/it, loss=0.0781, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1342/2855 [50:39<42:07,  1.67s/it, loss=0.0781, lr=2.77e-05]
Epoch 2/8:  47%|████▋     | 1342/2855 [50:41<42:07,  1.67s/it, loss=0.2892, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1343/2855 [50:41<40:40,  1.61s/it, loss=0.2892, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1343/2855 [50:42<40:40,  1.61s/it, loss=0.2763, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1344/2855 [50:42<41:53,  1.66s/it, loss=0.2763, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1344/2855 [50:44<41:53,  1.66s/it, loss=0.1958, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1345/2855 [50:44<40:50,  1.62s/it, loss=0.1958, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1345/2855 [50:45<40:50,  1.62s/it, loss=0.1675, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1346/2855 [50:45<39:21,  1.57s/it, loss=0.1675, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1346/2855 [50:47<39:21,  1.57s/it, loss=0.2258, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1347/2855 [50:47<40:36,  1.62s/it, loss=0.2258, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1347/2855 [50:49<40:36,  1.62s/it, loss=0.0615, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1348/2855 [50:49<40:46,  1.62s/it, loss=0.0615, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1348/2855 [50:50<40:46,  1.62s/it, loss=0.0667, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1349/2855 [50:50<40:21,  1.61s/it, loss=0.0667, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1349/2855 [50:52<40:21,  1.61s/it, loss=0.0985, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1350/2855 [50:52<42:04,  1.68s/it, loss=0.0985, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1350/2855 [50:54<42:04,  1.68s/it, loss=0.1055, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1351/2855 [50:54<40:42,  1.62s/it, loss=0.1055, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1351/2855 [50:55<40:42,  1.62s/it, loss=0.0852, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1352/2855 [50:55<38:55,  1.55s/it, loss=0.0852, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1352/2855 [50:57<38:55,  1.55s/it, loss=0.1536, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1353/2855 [50:57<39:40,  1.58s/it, loss=0.1536, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1353/2855 [50:58<39:40,  1.58s/it, loss=0.0589, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1354/2855 [50:58<39:09,  1.57s/it, loss=0.0589, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1354/2855 [51:00<39:09,  1.57s/it, loss=0.1080, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1355/2855 [51:00<38:03,  1.52s/it, loss=0.1080, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1355/2855 [51:01<38:03,  1.52s/it, loss=0.1453, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1356/2855 [51:01<38:25,  1.54s/it, loss=0.1453, lr=2.76e-05]
Epoch 2/8:  47%|████▋     | 1356/2855 [51:03<38:25,  1.54s/it, loss=0.1776, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1357/2855 [51:03<38:53,  1.56s/it, loss=0.1776, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1357/2855 [51:04<38:53,  1.56s/it, loss=0.0854, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1358/2855 [51:04<37:59,  1.52s/it, loss=0.0854, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1358/2855 [51:06<37:59,  1.52s/it, loss=0.1686, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1359/2855 [51:06<37:13,  1.49s/it, loss=0.1686, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1359/2855 [51:07<37:13,  1.49s/it, loss=0.2397, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1360/2855 [51:07<39:08,  1.57s/it, loss=0.2397, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1360/2855 [51:09<39:08,  1.57s/it, loss=0.1024, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1361/2855 [51:09<37:55,  1.52s/it, loss=0.1024, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1361/2855 [51:10<37:55,  1.52s/it, loss=0.0889, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1362/2855 [51:10<35:55,  1.44s/it, loss=0.0889, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1362/2855 [51:11<35:55,  1.44s/it, loss=0.2483, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1363/2855 [51:11<35:44,  1.44s/it, loss=0.2483, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1363/2855 [51:13<35:44,  1.44s/it, loss=0.1279, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1364/2855 [51:13<35:21,  1.42s/it, loss=0.1279, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1364/2855 [51:14<35:21,  1.42s/it, loss=0.1640, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1365/2855 [51:14<34:23,  1.39s/it, loss=0.1640, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1365/2855 [51:15<34:23,  1.39s/it, loss=0.2754, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1366/2855 [51:15<33:25,  1.35s/it, loss=0.2754, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1366/2855 [51:17<33:25,  1.35s/it, loss=0.0776, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1367/2855 [51:17<34:17,  1.38s/it, loss=0.0776, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1367/2855 [51:18<34:17,  1.38s/it, loss=0.1532, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1368/2855 [51:18<34:04,  1.37s/it, loss=0.1532, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1368/2855 [51:20<34:04,  1.37s/it, loss=0.1649, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1369/2855 [51:20<33:24,  1.35s/it, loss=0.1649, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1369/2855 [51:21<33:24,  1.35s/it, loss=0.1152, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1370/2855 [51:21<32:37,  1.32s/it, loss=0.1152, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1370/2855 [51:22<32:37,  1.32s/it, loss=0.1445, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1371/2855 [51:22<33:56,  1.37s/it, loss=0.1445, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1371/2855 [51:24<33:56,  1.37s/it, loss=0.2108, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1372/2855 [51:24<33:56,  1.37s/it, loss=0.2108, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1372/2855 [51:25<33:56,  1.37s/it, loss=0.4202, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1373/2855 [51:25<33:21,  1.35s/it, loss=0.4202, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1373/2855 [51:26<33:21,  1.35s/it, loss=0.1523, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1374/2855 [51:26<33:01,  1.34s/it, loss=0.1523, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1374/2855 [51:28<33:01,  1.34s/it, loss=0.1135, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1375/2855 [51:28<34:59,  1.42s/it, loss=0.1135, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1375/2855 [51:29<34:59,  1.42s/it, loss=0.0984, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1376/2855 [51:29<34:23,  1.40s/it, loss=0.0984, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1376/2855 [51:31<34:23,  1.40s/it, loss=0.2992, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1377/2855 [51:31<34:40,  1.41s/it, loss=0.2992, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1377/2855 [51:32<34:40,  1.41s/it, loss=0.3621, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1378/2855 [51:32<37:20,  1.52s/it, loss=0.3621, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1378/2855 [51:34<37:20,  1.52s/it, loss=0.2765, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1379/2855 [51:34<37:10,  1.51s/it, loss=0.2765, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1379/2855 [51:35<37:10,  1.51s/it, loss=0.0533, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1380/2855 [51:35<37:41,  1.53s/it, loss=0.0533, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1380/2855 [51:37<37:41,  1.53s/it, loss=0.2031, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1381/2855 [51:37<39:45,  1.62s/it, loss=0.2031, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1381/2855 [51:39<39:45,  1.62s/it, loss=0.1681, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1382/2855 [51:39<38:56,  1.59s/it, loss=0.1681, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1382/2855 [51:40<38:56,  1.59s/it, loss=0.1460, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1383/2855 [51:40<37:34,  1.53s/it, loss=0.1460, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1383/2855 [51:42<37:34,  1.53s/it, loss=0.2899, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1384/2855 [51:42<38:09,  1.56s/it, loss=0.2899, lr=2.76e-05]
Epoch 2/8:  48%|████▊     | 1384/2855 [51:43<38:09,  1.56s/it, loss=0.1505, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1385/2855 [51:43<38:17,  1.56s/it, loss=0.1505, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1385/2855 [51:45<38:17,  1.56s/it, loss=0.2266, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1386/2855 [51:45<37:16,  1.52s/it, loss=0.2266, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1386/2855 [51:46<37:16,  1.52s/it, loss=0.1544, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1387/2855 [51:46<36:53,  1.51s/it, loss=0.1544, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1387/2855 [51:48<36:53,  1.51s/it, loss=0.1366, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1388/2855 [51:48<38:42,  1.58s/it, loss=0.1366, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1388/2855 [51:49<38:42,  1.58s/it, loss=0.0547, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1389/2855 [51:49<37:28,  1.53s/it, loss=0.0547, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1389/2855 [51:51<37:28,  1.53s/it, loss=0.2170, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1390/2855 [51:51<35:34,  1.46s/it, loss=0.2170, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1390/2855 [51:52<35:34,  1.46s/it, loss=0.1823, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1391/2855 [51:52<35:22,  1.45s/it, loss=0.1823, lr=2.76e-05]
Epoch 2/8:  49%|████▊     | 1391/2855 [51:54<35:22,  1.45s/it, loss=0.3025, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1392/2855 [51:54<34:14,  1.40s/it, loss=0.3025, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1392/2855 [51:55<34:14,  1.40s/it, loss=0.0809, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1393/2855 [51:55<33:07,  1.36s/it, loss=0.0809, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1393/2855 [51:56<33:07,  1.36s/it, loss=0.1499, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1394/2855 [51:56<32:16,  1.33s/it, loss=0.1499, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1394/2855 [51:57<32:16,  1.33s/it, loss=0.2533, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1395/2855 [51:57<33:23,  1.37s/it, loss=0.2533, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1395/2855 [51:59<33:23,  1.37s/it, loss=0.1995, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1396/2855 [51:59<32:55,  1.35s/it, loss=0.1995, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1396/2855 [52:00<32:55,  1.35s/it, loss=0.2100, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1397/2855 [52:00<32:01,  1.32s/it, loss=0.2100, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1397/2855 [52:01<32:01,  1.32s/it, loss=0.0641, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1398/2855 [52:01<31:36,  1.30s/it, loss=0.0641, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1398/2855 [52:03<31:36,  1.30s/it, loss=0.0798, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1399/2855 [52:03<32:53,  1.36s/it, loss=0.0798, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1399/2855 [52:04<32:53,  1.36s/it, loss=0.0876, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1400/2855 [52:04<32:32,  1.34s/it, loss=0.0876, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1400/2855 [52:05<32:32,  1.34s/it, loss=0.0667, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1401/2855 [52:05<31:47,  1.31s/it, loss=0.0667, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1401/2855 [52:07<31:47,  1.31s/it, loss=0.2626, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1402/2855 [52:07<31:58,  1.32s/it, loss=0.2626, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1402/2855 [52:08<31:58,  1.32s/it, loss=0.1477, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1403/2855 [52:08<33:13,  1.37s/it, loss=0.1477, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1403/2855 [52:10<33:13,  1.37s/it, loss=0.1391, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1404/2855 [52:10<34:16,  1.42s/it, loss=0.1391, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1404/2855 [52:11<34:16,  1.42s/it, loss=0.2052, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1405/2855 [52:11<34:56,  1.45s/it, loss=0.2052, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1405/2855 [52:13<34:56,  1.45s/it, loss=0.1692, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1406/2855 [52:13<37:07,  1.54s/it, loss=0.1692, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1406/2855 [52:14<37:07,  1.54s/it, loss=0.2051, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1407/2855 [52:14<35:52,  1.49s/it, loss=0.2051, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1407/2855 [52:16<35:52,  1.49s/it, loss=0.1378, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1408/2855 [52:16<34:06,  1.41s/it, loss=0.1378, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1408/2855 [52:17<34:06,  1.41s/it, loss=0.0624, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1409/2855 [52:17<36:32,  1.52s/it, loss=0.0624, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1409/2855 [1:01:08<36:32,  1.52s/it, loss=0.1276, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1410/2855 [1:01:08<64:21:28, 160.34s/it, loss=0.1276, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1410/2855 [1:01:11<64:21:28, 160.34s/it, loss=0.1108, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1411/2855 [1:01:11<45:17:50, 112.93s/it, loss=0.1108, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1411/2855 [1:01:12<45:17:50, 112.93s/it, loss=0.1889, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1412/2855 [1:01:12<31:50:37, 79.44s/it, loss=0.1889, lr=2.76e-05] 
Epoch 2/8:  49%|████▉     | 1412/2855 [1:01:13<31:50:37, 79.44s/it, loss=0.1159, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1413/2855 [1:01:13<22:26:36, 56.03s/it, loss=0.1159, lr=2.76e-05]
Epoch 2/8:  49%|████▉     | 1413/2855 [1:01:18<22:26:36, 56.03s/it, loss=0.1760, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1414/2855 [1:01:18<16:16:36, 40.66s/it, loss=0.1760, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1414/2855 [1:01:19<16:16:36, 40.66s/it, loss=0.0752, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1415/2855 [1:01:19<11:31:51, 28.83s/it, loss=0.0752, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1415/2855 [1:01:20<11:31:51, 28.83s/it, loss=0.1296, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1416/2855 [1:01:20<8:12:39, 20.54s/it, loss=0.1296, lr=2.76e-05] 
Epoch 2/8:  50%|████▉     | 1416/2855 [1:01:22<8:12:39, 20.54s/it, loss=0.1767, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1417/2855 [1:01:22<5:53:30, 14.75s/it, loss=0.1767, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1417/2855 [1:01:23<5:53:30, 14.75s/it, loss=0.0897, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1418/2855 [1:01:23<4:16:47, 10.72s/it, loss=0.0897, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1418/2855 [1:01:24<4:16:47, 10.72s/it, loss=0.1338, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1419/2855 [1:01:24<3:09:24,  7.91s/it, loss=0.1338, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1419/2855 [1:01:26<3:09:24,  7.91s/it, loss=0.2463, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1420/2855 [1:01:26<2:21:27,  5.91s/it, loss=0.2463, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1420/2855 [1:01:27<2:21:27,  5.91s/it, loss=0.2160, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1421/2855 [1:01:27<1:48:22,  4.53s/it, loss=0.2160, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1421/2855 [1:01:28<1:48:22,  4.53s/it, loss=0.0725, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1422/2855 [1:01:28<1:25:57,  3.60s/it, loss=0.0725, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1422/2855 [1:01:30<1:25:57,  3.60s/it, loss=0.2019, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1423/2855 [1:01:30<1:09:41,  2.92s/it, loss=0.2019, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1423/2855 [1:01:31<1:09:41,  2.92s/it, loss=0.0966, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1424/2855 [1:01:31<58:27,  2.45s/it, loss=0.0966, lr=2.76e-05]  
Epoch 2/8:  50%|████▉     | 1424/2855 [1:01:32<58:27,  2.45s/it, loss=0.1850, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1425/2855 [1:01:32<50:38,  2.12s/it, loss=0.1850, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1425/2855 [1:01:34<50:38,  2.12s/it, loss=0.1061, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1426/2855 [1:01:34<46:00,  1.93s/it, loss=0.1061, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1426/2855 [1:01:35<46:00,  1.93s/it, loss=0.0740, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1427/2855 [1:01:35<41:39,  1.75s/it, loss=0.0740, lr=2.76e-05]
Epoch 2/8:  50%|████▉     | 1427/2855 [1:01:37<41:39,  1.75s/it, loss=0.0378, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1428/2855 [1:01:37<38:41,  1.63s/it, loss=0.0378, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1428/2855 [1:01:38<38:41,  1.63s/it, loss=0.2611, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1429/2855 [1:01:38<37:10,  1.56s/it, loss=0.2611, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1429/2855 [1:01:39<37:10,  1.56s/it, loss=0.0890, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1430/2855 [1:01:39<35:56,  1.51s/it, loss=0.0890, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1430/2855 [1:01:41<35:56,  1.51s/it, loss=0.1275, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1431/2855 [1:01:41<34:54,  1.47s/it, loss=0.1275, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1431/2855 [1:01:42<34:54,  1.47s/it, loss=0.1214, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1432/2855 [1:01:42<34:31,  1.46s/it, loss=0.1214, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1432/2855 [1:01:44<34:31,  1.46s/it, loss=0.4741, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1433/2855 [1:01:44<35:02,  1.48s/it, loss=0.4741, lr=2.76e-05]
Epoch 2/8:  50%|█████     | 1433/2855 [1:01:45<35:02,  1.48s/it, loss=0.0539, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1434/2855 [1:01:45<34:18,  1.45s/it, loss=0.0539, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1434/2855 [1:01:46<34:18,  1.45s/it, loss=0.0814, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1435/2855 [1:01:46<33:45,  1.43s/it, loss=0.0814, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1435/2855 [1:01:48<33:45,  1.43s/it, loss=0.1275, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1436/2855 [1:01:48<34:34,  1.46s/it, loss=0.1275, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1436/2855 [1:01:50<34:34,  1.46s/it, loss=0.1366, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1437/2855 [1:01:50<34:55,  1.48s/it, loss=0.1366, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1437/2855 [1:01:51<34:55,  1.48s/it, loss=0.1106, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1438/2855 [1:01:51<34:02,  1.44s/it, loss=0.1106, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1438/2855 [1:01:52<34:02,  1.44s/it, loss=0.2052, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1439/2855 [1:01:52<33:37,  1.42s/it, loss=0.2052, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1439/2855 [1:01:54<33:37,  1.42s/it, loss=0.1972, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1440/2855 [1:01:54<34:33,  1.47s/it, loss=0.1972, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1440/2855 [1:01:55<34:33,  1.47s/it, loss=0.1830, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1441/2855 [1:01:55<33:49,  1.44s/it, loss=0.1830, lr=2.75e-05]
Epoch 2/8:  50%|█████     | 1441/2855 [1:01:57<33:49,  1.44s/it, loss=0.1388, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1442/2855 [1:01:57<33:15,  1.41s/it, loss=0.1388, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1442/2855 [1:01:58<33:15,  1.41s/it, loss=0.1680, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1443/2855 [1:01:58<33:45,  1.43s/it, loss=0.1680, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1443/2855 [1:02:00<33:45,  1.43s/it, loss=0.1531, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1444/2855 [1:02:00<34:27,  1.47s/it, loss=0.1531, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1444/2855 [1:02:01<34:27,  1.47s/it, loss=0.1999, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1445/2855 [1:02:01<33:47,  1.44s/it, loss=0.1999, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1445/2855 [1:02:02<33:47,  1.44s/it, loss=0.1140, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1446/2855 [1:02:02<33:41,  1.43s/it, loss=0.1140, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1446/2855 [1:02:04<33:41,  1.43s/it, loss=0.2247, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1447/2855 [1:02:04<35:38,  1.52s/it, loss=0.2247, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1447/2855 [1:02:06<35:38,  1.52s/it, loss=0.2798, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1448/2855 [1:02:06<35:29,  1.51s/it, loss=0.2798, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1448/2855 [1:02:07<35:29,  1.51s/it, loss=0.0309, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1449/2855 [1:02:07<35:04,  1.50s/it, loss=0.0309, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1449/2855 [1:02:09<35:04,  1.50s/it, loss=0.1445, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1450/2855 [1:02:09<36:18,  1.55s/it, loss=0.1445, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1450/2855 [1:02:10<36:18,  1.55s/it, loss=0.0660, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1451/2855 [1:02:10<36:04,  1.54s/it, loss=0.0660, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1451/2855 [1:02:12<36:04,  1.54s/it, loss=0.1311, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1452/2855 [1:02:12<35:07,  1.50s/it, loss=0.1311, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1452/2855 [1:02:13<35:07,  1.50s/it, loss=0.1426, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1453/2855 [1:02:13<35:01,  1.50s/it, loss=0.1426, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1453/2855 [1:02:16<35:01,  1.50s/it, loss=0.1873, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1454/2855 [1:02:16<43:13,  1.85s/it, loss=0.1873, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1454/2855 [1:02:17<43:13,  1.85s/it, loss=0.2453, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1455/2855 [1:02:17<40:24,  1.73s/it, loss=0.2453, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1455/2855 [1:02:19<40:24,  1.73s/it, loss=0.0987, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1456/2855 [1:02:19<39:11,  1.68s/it, loss=0.0987, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1456/2855 [1:02:20<39:11,  1.68s/it, loss=0.1542, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1457/2855 [1:02:20<38:27,  1.65s/it, loss=0.1542, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1457/2855 [1:02:22<38:27,  1.65s/it, loss=0.2307, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1458/2855 [1:02:22<38:10,  1.64s/it, loss=0.2307, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1458/2855 [1:02:24<38:10,  1.64s/it, loss=0.1829, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1459/2855 [1:02:24<40:16,  1.73s/it, loss=0.1829, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1459/2855 [1:02:26<40:16,  1.73s/it, loss=0.2257, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1460/2855 [1:02:26<39:21,  1.69s/it, loss=0.2257, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1460/2855 [1:02:27<39:21,  1.69s/it, loss=0.1014, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1461/2855 [1:02:27<38:18,  1.65s/it, loss=0.1014, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1461/2855 [1:02:29<38:18,  1.65s/it, loss=0.1666, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1462/2855 [1:02:29<39:46,  1.71s/it, loss=0.1666, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1462/2855 [1:02:31<39:46,  1.71s/it, loss=0.1391, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1463/2855 [1:02:31<40:34,  1.75s/it, loss=0.1391, lr=2.75e-05]
Epoch 2/8:  51%|█████     | 1463/2855 [1:02:32<40:34,  1.75s/it, loss=0.0704, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1464/2855 [1:02:32<39:33,  1.71s/it, loss=0.0704, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1464/2855 [1:02:34<39:33,  1.71s/it, loss=0.2147, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1465/2855 [1:02:34<40:59,  1.77s/it, loss=0.2147, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1465/2855 [1:02:36<40:59,  1.77s/it, loss=0.2686, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1466/2855 [1:02:36<40:12,  1.74s/it, loss=0.2686, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1466/2855 [1:02:38<40:12,  1.74s/it, loss=0.3187, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1467/2855 [1:02:38<39:03,  1.69s/it, loss=0.3187, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1467/2855 [1:02:40<39:03,  1.69s/it, loss=0.0494, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1468/2855 [1:02:40<40:38,  1.76s/it, loss=0.0494, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1468/2855 [1:02:41<40:38,  1.76s/it, loss=0.0186, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1469/2855 [1:02:41<40:00,  1.73s/it, loss=0.0186, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1469/2855 [1:02:43<40:00,  1.73s/it, loss=0.0386, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1470/2855 [1:02:43<39:56,  1.73s/it, loss=0.0386, lr=2.75e-05]
Epoch 2/8:  51%|█████▏    | 1470/2855 [1:02:45<39:56,  1.73s/it, loss=0.1550, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1471/2855 [1:02:45<42:01,  1.82s/it, loss=0.1550, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1471/2855 [1:02:47<42:01,  1.82s/it, loss=0.2588, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1472/2855 [1:02:47<41:03,  1.78s/it, loss=0.2588, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1472/2855 [1:02:48<41:03,  1.78s/it, loss=0.1123, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1473/2855 [1:02:48<40:37,  1.76s/it, loss=0.1123, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1473/2855 [1:02:51<40:37,  1.76s/it, loss=0.1857, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1474/2855 [1:02:51<44:50,  1.95s/it, loss=0.1857, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1474/2855 [1:02:53<44:50,  1.95s/it, loss=0.2820, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1475/2855 [1:02:53<43:56,  1.91s/it, loss=0.2820, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1475/2855 [1:02:55<43:56,  1.91s/it, loss=0.2588, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1476/2855 [1:02:55<46:52,  2.04s/it, loss=0.2588, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1476/2855 [1:02:57<46:52,  2.04s/it, loss=0.1286, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1477/2855 [1:02:57<50:42,  2.21s/it, loss=0.1286, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1477/2855 [1:03:00<50:42,  2.21s/it, loss=0.2075, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1478/2855 [1:03:00<51:11,  2.23s/it, loss=0.2075, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1478/2855 [1:03:02<51:11,  2.23s/it, loss=0.0835, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1479/2855 [1:03:02<49:16,  2.15s/it, loss=0.0835, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1479/2855 [1:03:04<49:16,  2.15s/it, loss=0.1884, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1480/2855 [1:03:04<47:35,  2.08s/it, loss=0.1884, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1480/2855 [1:03:06<47:35,  2.08s/it, loss=0.2436, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1481/2855 [1:03:06<47:44,  2.09s/it, loss=0.2436, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1481/2855 [1:03:08<47:44,  2.09s/it, loss=0.1178, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1482/2855 [1:03:08<45:54,  2.01s/it, loss=0.1178, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1482/2855 [1:03:10<45:54,  2.01s/it, loss=0.1410, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1483/2855 [1:03:10<45:40,  2.00s/it, loss=0.1410, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1483/2855 [1:03:11<45:40,  2.00s/it, loss=0.1899, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1484/2855 [1:03:11<44:12,  1.94s/it, loss=0.1899, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1484/2855 [1:03:13<44:12,  1.94s/it, loss=0.2049, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1485/2855 [1:03:13<43:44,  1.92s/it, loss=0.2049, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1485/2855 [1:03:15<43:44,  1.92s/it, loss=0.2670, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1486/2855 [1:03:15<44:14,  1.94s/it, loss=0.2670, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1486/2855 [1:03:17<44:14,  1.94s/it, loss=0.1557, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1487/2855 [1:03:17<43:58,  1.93s/it, loss=0.1557, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1487/2855 [1:03:19<43:58,  1.93s/it, loss=0.0854, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1488/2855 [1:03:19<43:36,  1.91s/it, loss=0.0854, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1488/2855 [1:03:21<43:36,  1.91s/it, loss=0.0579, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1489/2855 [1:03:21<42:44,  1.88s/it, loss=0.0579, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1489/2855 [1:03:22<42:44,  1.88s/it, loss=0.1763, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1490/2855 [1:03:22<41:25,  1.82s/it, loss=0.1763, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1490/2855 [1:03:24<41:25,  1.82s/it, loss=0.1688, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1491/2855 [1:03:24<41:44,  1.84s/it, loss=0.1688, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1491/2855 [1:03:26<41:44,  1.84s/it, loss=0.0872, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1492/2855 [1:03:26<41:13,  1.81s/it, loss=0.0872, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1492/2855 [1:03:33<41:13,  1.81s/it, loss=0.2936, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1493/2855 [1:03:33<1:16:46,  3.38s/it, loss=0.2936, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1493/2855 [1:03:35<1:16:46,  3.38s/it, loss=0.1746, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1494/2855 [1:03:35<1:03:34,  2.80s/it, loss=0.1746, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1494/2855 [1:03:36<1:03:34,  2.80s/it, loss=0.2125, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1495/2855 [1:03:36<56:14,  2.48s/it, loss=0.2125, lr=2.75e-05]  
Epoch 2/8:  52%|█████▏    | 1495/2855 [1:03:38<56:14,  2.48s/it, loss=0.0639, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1496/2855 [1:03:38<53:38,  2.37s/it, loss=0.0639, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1496/2855 [1:03:40<53:38,  2.37s/it, loss=0.0134, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1497/2855 [1:03:40<50:18,  2.22s/it, loss=0.0134, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1497/2855 [1:03:42<50:18,  2.22s/it, loss=0.1632, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1498/2855 [1:03:42<45:14,  2.00s/it, loss=0.1632, lr=2.75e-05]
Epoch 2/8:  52%|█████▏    | 1498/2855 [1:03:43<45:14,  2.00s/it, loss=0.1279, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1499/2855 [1:03:43<41:15,  1.83s/it, loss=0.1279, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1499/2855 [1:03:45<41:15,  1.83s/it, loss=0.2589, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1500/2855 [1:03:45<41:21,  1.83s/it, loss=0.2589, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1500/2855 [1:03:47<41:21,  1.83s/it, loss=0.2867, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1501/2855 [1:03:47<41:17,  1.83s/it, loss=0.2867, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1501/2855 [1:03:49<41:17,  1.83s/it, loss=0.0678, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1502/2855 [1:03:49<40:01,  1.77s/it, loss=0.0678, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1502/2855 [1:03:50<40:01,  1.77s/it, loss=0.3284, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1503/2855 [1:03:50<41:00,  1.82s/it, loss=0.3284, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1503/2855 [1:03:52<41:00,  1.82s/it, loss=0.1686, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1504/2855 [1:03:52<40:01,  1.78s/it, loss=0.1686, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1504/2855 [1:03:54<40:01,  1.78s/it, loss=0.0884, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1505/2855 [1:03:54<39:16,  1.75s/it, loss=0.0884, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1505/2855 [1:03:56<39:16,  1.75s/it, loss=0.1195, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1506/2855 [1:03:56<39:47,  1.77s/it, loss=0.1195, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1506/2855 [1:03:57<39:47,  1.77s/it, loss=0.1542, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1507/2855 [1:03:57<38:53,  1.73s/it, loss=0.1542, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1507/2855 [1:03:59<38:53,  1.73s/it, loss=0.0609, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1508/2855 [1:03:59<38:47,  1.73s/it, loss=0.0609, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1508/2855 [1:04:01<38:47,  1.73s/it, loss=0.0705, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1509/2855 [1:04:01<39:38,  1.77s/it, loss=0.0705, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1509/2855 [1:04:02<39:38,  1.77s/it, loss=0.1707, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1510/2855 [1:04:02<38:08,  1.70s/it, loss=0.1707, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1510/2855 [1:04:04<38:08,  1.70s/it, loss=0.1740, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1511/2855 [1:04:04<37:06,  1.66s/it, loss=0.1740, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1511/2855 [1:04:06<37:06,  1.66s/it, loss=0.2363, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1512/2855 [1:04:06<39:34,  1.77s/it, loss=0.2363, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1512/2855 [1:04:08<39:34,  1.77s/it, loss=0.1774, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1513/2855 [1:04:08<40:58,  1.83s/it, loss=0.1774, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1513/2855 [1:04:10<40:58,  1.83s/it, loss=0.0866, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1514/2855 [1:04:10<41:30,  1.86s/it, loss=0.0866, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1514/2855 [1:04:12<41:30,  1.86s/it, loss=0.1028, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1515/2855 [1:04:12<42:14,  1.89s/it, loss=0.1028, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1515/2855 [1:04:14<42:14,  1.89s/it, loss=0.2351, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1516/2855 [1:04:14<41:13,  1.85s/it, loss=0.2351, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1516/2855 [1:04:16<41:13,  1.85s/it, loss=0.0751, lr=2.75e-05]
Epoch 2/8:  53%|████��▎    | 1517/2855 [1:04:16<41:37,  1.87s/it, loss=0.0751, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1517/2855 [1:04:17<41:37,  1.87s/it, loss=0.2863, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1518/2855 [1:04:17<41:12,  1.85s/it, loss=0.2863, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1518/2855 [1:04:19<41:12,  1.85s/it, loss=0.2365, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1519/2855 [1:04:19<40:39,  1.83s/it, loss=0.2365, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1519/2855 [1:04:21<40:39,  1.83s/it, loss=0.1284, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1520/2855 [1:04:21<41:42,  1.87s/it, loss=0.1284, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1520/2855 [1:04:23<41:42,  1.87s/it, loss=0.1473, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1521/2855 [1:04:23<40:03,  1.80s/it, loss=0.1473, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1521/2855 [1:04:24<40:03,  1.80s/it, loss=0.1167, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1522/2855 [1:04:24<38:56,  1.75s/it, loss=0.1167, lr=2.75e-05]
Epoch 2/8:  53%|█████▎    | 1522/2855 [1:04:26<38:56,  1.75s/it, loss=0.0931, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1523/2855 [1:04:26<38:47,  1.75s/it, loss=0.0931, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1523/2855 [1:04:28<38:47,  1.75s/it, loss=0.0876, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1524/2855 [1:04:28<37:31,  1.69s/it, loss=0.0876, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1524/2855 [1:04:29<37:31,  1.69s/it, loss=0.1276, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1525/2855 [1:04:29<37:43,  1.70s/it, loss=0.1276, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1525/2855 [1:04:31<37:43,  1.70s/it, loss=0.0896, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1526/2855 [1:04:31<39:05,  1.76s/it, loss=0.0896, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1526/2855 [1:04:33<39:05,  1.76s/it, loss=0.0833, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1527/2855 [1:04:33<38:22,  1.73s/it, loss=0.0833, lr=2.74e-05]
Epoch 2/8:  53%|█████▎    | 1527/2855 [1:04:34<38:22,  1.73s/it, loss=0.0379, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1528/2855 [1:04:34<37:07,  1.68s/it, loss=0.0379, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1528/2855 [1:04:36<37:07,  1.68s/it, loss=0.0221, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1529/2855 [1:04:36<37:35,  1.70s/it, loss=0.0221, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1529/2855 [1:04:38<37:35,  1.70s/it, loss=0.1383, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1530/2855 [1:04:38<38:02,  1.72s/it, loss=0.1383, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1530/2855 [1:04:40<38:02,  1.72s/it, loss=0.1545, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1531/2855 [1:04:40<37:58,  1.72s/it, loss=0.1545, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1531/2855 [1:04:41<37:58,  1.72s/it, loss=0.1252, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1532/2855 [1:04:41<37:52,  1.72s/it, loss=0.1252, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1532/2855 [1:04:47<37:52,  1.72s/it, loss=0.0906, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1533/2855 [1:04:47<1:01:46,  2.80s/it, loss=0.0906, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1533/2855 [1:04:48<1:01:46,  2.80s/it, loss=0.0426, lr=2.74e-05]
Epoch 2/8:  54%|█████▎    | 1534/2855 [1:04:48<51:43,  2.35s/it, loss=0.0426, lr=2.74e-05]  
Epoch 2/8:  54%|█████▎    | 1534/2855 [1:04:50<51:43,  2.35s/it, loss=0.2004, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1535/2855 [1:04:50<46:16,  2.10s/it, loss=0.2004, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1535/2855 [1:04:52<46:16,  2.10s/it, loss=0.2266, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1536/2855 [1:04:52<45:11,  2.06s/it, loss=0.2266, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1536/2855 [1:04:53<45:11,  2.06s/it, loss=0.2050, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1537/2855 [1:04:53<43:17,  1.97s/it, loss=0.2050, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1537/2855 [1:04:55<43:17,  1.97s/it, loss=0.2134, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1538/2855 [1:04:55<41:13,  1.88s/it, loss=0.2134, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1538/2855 [1:04:57<41:13,  1.88s/it, loss=0.2391, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1539/2855 [1:04:57<39:38,  1.81s/it, loss=0.2391, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1539/2855 [1:04:58<39:38,  1.81s/it, loss=0.0356, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1540/2855 [1:04:58<38:15,  1.75s/it, loss=0.0356, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1540/2855 [1:05:00<38:15,  1.75s/it, loss=0.0888, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1541/2855 [1:05:00<37:04,  1.69s/it, loss=0.0888, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1541/2855 [1:05:02<37:04,  1.69s/it, loss=0.2068, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1542/2855 [1:05:02<37:18,  1.71s/it, loss=0.2068, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1542/2855 [1:05:03<37:18,  1.71s/it, loss=0.1665, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1543/2855 [1:05:03<38:06,  1.74s/it, loss=0.1665, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1543/2855 [1:05:05<38:06,  1.74s/it, loss=0.1223, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1544/2855 [1:05:05<38:47,  1.78s/it, loss=0.1223, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1544/2855 [1:05:07<38:47,  1.78s/it, loss=0.0459, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1545/2855 [1:05:07<38:46,  1.78s/it, loss=0.0459, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1545/2855 [1:05:09<38:46,  1.78s/it, loss=0.1723, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1546/2855 [1:05:09<39:35,  1.81s/it, loss=0.1723, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1546/2855 [1:05:11<39:35,  1.81s/it, loss=0.1445, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1547/2855 [1:05:11<38:34,  1.77s/it, loss=0.1445, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1547/2855 [1:05:12<38:34,  1.77s/it, loss=0.0748, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1548/2855 [1:05:12<38:42,  1.78s/it, loss=0.0748, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1548/2855 [1:05:14<38:42,  1.78s/it, loss=0.1235, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1549/2855 [1:05:14<38:09,  1.75s/it, loss=0.1235, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1549/2855 [1:05:16<38:09,  1.75s/it, loss=0.1357, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1550/2855 [1:05:16<38:55,  1.79s/it, loss=0.1357, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1550/2855 [1:05:18<38:55,  1.79s/it, loss=0.1631, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1551/2855 [1:05:18<38:29,  1.77s/it, loss=0.1631, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1551/2855 [1:05:19<38:29,  1.77s/it, loss=0.1668, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1552/2855 [1:05:19<37:24,  1.72s/it, loss=0.1668, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1552/2855 [1:05:21<37:24,  1.72s/it, loss=0.2598, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1553/2855 [1:05:21<37:22,  1.72s/it, loss=0.2598, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1553/2855 [1:05:23<37:22,  1.72s/it, loss=0.1346, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1554/2855 [1:05:23<37:49,  1.74s/it, loss=0.1346, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1554/2855 [1:05:25<37:49,  1.74s/it, loss=0.1023, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1555/2855 [1:05:25<38:24,  1.77s/it, loss=0.1023, lr=2.74e-05]
Epoch 2/8:  54%|█████▍    | 1555/2855 [1:05:26<38:24,  1.77s/it, loss=0.1328, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1556/2855 [1:05:26<38:58,  1.80s/it, loss=0.1328, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1556/2855 [1:05:28<38:58,  1.80s/it, loss=0.2296, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1557/2855 [1:05:28<38:42,  1.79s/it, loss=0.2296, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1557/2855 [1:05:30<38:42,  1.79s/it, loss=0.1597, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1558/2855 [1:05:30<38:35,  1.79s/it, loss=0.1597, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1558/2855 [1:05:32<38:35,  1.79s/it, loss=0.1140, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1559/2855 [1:05:32<38:45,  1.79s/it, loss=0.1140, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1559/2855 [1:05:34<38:45,  1.79s/it, loss=0.1759, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1560/2855 [1:05:34<38:12,  1.77s/it, loss=0.1759, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1560/2855 [1:05:35<38:12,  1.77s/it, loss=0.1753, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1561/2855 [1:05:35<37:41,  1.75s/it, loss=0.1753, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1561/2855 [1:05:37<37:41,  1.75s/it, loss=0.1555, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1562/2855 [1:05:37<37:58,  1.76s/it, loss=0.1555, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1562/2855 [1:05:39<37:58,  1.76s/it, loss=0.1553, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1563/2855 [1:05:39<37:24,  1.74s/it, loss=0.1553, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1563/2855 [1:05:40<37:24,  1.74s/it, loss=0.1091, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1564/2855 [1:05:40<37:10,  1.73s/it, loss=0.1091, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1564/2855 [1:05:42<37:10,  1.73s/it, loss=0.1320, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1565/2855 [1:05:42<37:20,  1.74s/it, loss=0.1320, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1565/2855 [1:05:44<37:20,  1.74s/it, loss=0.1797, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1566/2855 [1:05:44<38:08,  1.78s/it, loss=0.1797, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1566/2855 [1:05:46<38:08,  1.78s/it, loss=0.1140, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1567/2855 [1:05:46<37:48,  1.76s/it, loss=0.1140, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1567/2855 [1:05:48<37:48,  1.76s/it, loss=0.1399, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1568/2855 [1:05:48<39:21,  1.84s/it, loss=0.1399, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1568/2855 [1:05:50<39:21,  1.84s/it, loss=0.2061, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1569/2855 [1:05:50<38:34,  1.80s/it, loss=0.2061, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1569/2855 [1:05:51<38:34,  1.80s/it, loss=0.1403, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1570/2855 [1:05:51<38:33,  1.80s/it, loss=0.1403, lr=2.74e-05]
Epoch 2/8:  55%|█████▍    | 1570/2855 [1:05:53<38:33,  1.80s/it, loss=0.1880, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1571/2855 [1:05:53<38:35,  1.80s/it, loss=0.1880, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1571/2855 [1:05:55<38:35,  1.80s/it, loss=0.0825, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1572/2855 [1:05:55<37:39,  1.76s/it, loss=0.0825, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1572/2855 [1:06:01<37:39,  1.76s/it, loss=0.0349, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1573/2855 [1:06:01<1:03:31,  2.97s/it, loss=0.0349, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1573/2855 [1:06:02<1:03:31,  2.97s/it, loss=0.1186, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1574/2855 [1:06:02<55:02,  2.58s/it, loss=0.1186, lr=2.74e-05]  
Epoch 2/8:  55%|█████▌    | 1574/2855 [1:06:04<55:02,  2.58s/it, loss=0.1109, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1575/2855 [1:06:04<50:56,  2.39s/it, loss=0.1109, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1575/2855 [1:06:06<50:56,  2.39s/it, loss=0.1125, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1576/2855 [1:06:06<46:40,  2.19s/it, loss=0.1125, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1576/2855 [1:06:08<46:40,  2.19s/it, loss=0.2695, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1577/2855 [1:06:08<43:41,  2.05s/it, loss=0.2695, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1577/2855 [1:06:09<43:41,  2.05s/it, loss=0.2775, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1578/2855 [1:06:09<41:22,  1.94s/it, loss=0.2775, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1578/2855 [1:06:11<41:22,  1.94s/it, loss=0.2011, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1579/2855 [1:06:11<40:11,  1.89s/it, loss=0.2011, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1579/2855 [1:06:13<40:11,  1.89s/it, loss=0.1171, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1580/2855 [1:06:13<39:54,  1.88s/it, loss=0.1171, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1580/2855 [1:06:15<39:54,  1.88s/it, loss=0.2026, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1581/2855 [1:06:15<39:57,  1.88s/it, loss=0.2026, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1581/2855 [1:06:17<39:57,  1.88s/it, loss=0.0920, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1582/2855 [1:06:17<39:40,  1.87s/it, loss=0.0920, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1582/2855 [1:06:18<39:40,  1.87s/it, loss=0.0605, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1583/2855 [1:06:18<39:04,  1.84s/it, loss=0.0605, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1583/2855 [1:06:20<39:04,  1.84s/it, loss=0.2639, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1584/2855 [1:06:20<38:06,  1.80s/it, loss=0.2639, lr=2.74e-05]
Epoch 2/8:  55%|█████▌    | 1584/2855 [1:06:22<38:06,  1.80s/it, loss=0.1261, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1585/2855 [1:06:22<38:37,  1.82s/it, loss=0.1261, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1585/2855 [1:06:24<38:37,  1.82s/it, loss=0.2664, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1586/2855 [1:06:24<38:30,  1.82s/it, loss=0.2664, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1586/2855 [1:06:25<38:30,  1.82s/it, loss=0.1122, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1587/2855 [1:06:25<37:21,  1.77s/it, loss=0.1122, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1587/2855 [1:06:27<37:21,  1.77s/it, loss=0.1544, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1588/2855 [1:06:27<38:18,  1.81s/it, loss=0.1544, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1588/2855 [1:06:29<38:18,  1.81s/it, loss=0.1327, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1589/2855 [1:06:29<38:18,  1.82s/it, loss=0.1327, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1589/2855 [1:06:31<38:18,  1.82s/it, loss=0.2187, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1590/2855 [1:06:31<38:01,  1.80s/it, loss=0.2187, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1590/2855 [1:06:33<38:01,  1.80s/it, loss=0.0857, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1591/2855 [1:06:33<37:37,  1.79s/it, loss=0.0857, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1591/2855 [1:06:34<37:37,  1.79s/it, loss=0.2978, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1592/2855 [1:06:34<36:36,  1.74s/it, loss=0.2978, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1592/2855 [1:06:36<36:36,  1.74s/it, loss=0.0431, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1593/2855 [1:06:36<36:03,  1.71s/it, loss=0.0431, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1593/2855 [1:06:38<36:03,  1.71s/it, loss=0.2411, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1594/2855 [1:06:38<36:34,  1.74s/it, loss=0.2411, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1594/2855 [1:06:40<36:34,  1.74s/it, loss=0.2263, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1595/2855 [1:06:40<36:32,  1.74s/it, loss=0.2263, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1595/2855 [1:06:41<36:32,  1.74s/it, loss=0.1017, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1596/2855 [1:06:41<35:38,  1.70s/it, loss=0.1017, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1596/2855 [1:06:43<35:38,  1.70s/it, loss=0.2586, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1597/2855 [1:06:43<34:47,  1.66s/it, loss=0.2586, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1597/2855 [1:06:44<34:47,  1.66s/it, loss=0.2897, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1598/2855 [1:06:44<34:24,  1.64s/it, loss=0.2897, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1598/2855 [1:06:46<34:24,  1.64s/it, loss=0.0974, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1599/2855 [1:06:46<34:11,  1.63s/it, loss=0.0974, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1599/2855 [1:06:48<34:11,  1.63s/it, loss=0.0974, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1600/2855 [1:06:48<36:32,  1.75s/it, loss=0.0974, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1600/2855 [1:06:50<36:32,  1.75s/it, loss=0.1992, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1601/2855 [1:06:50<35:42,  1.71s/it, loss=0.1992, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1601/2855 [1:06:51<35:42,  1.71s/it, loss=0.2066, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1602/2855 [1:06:51<34:07,  1.63s/it, loss=0.2066, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1602/2855 [1:06:53<34:07,  1.63s/it, loss=0.0979, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1603/2855 [1:06:53<34:50,  1.67s/it, loss=0.0979, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1603/2855 [1:06:54<34:50,  1.67s/it, loss=0.0621, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1604/2855 [1:06:54<34:37,  1.66s/it, loss=0.0621, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1604/2855 [1:06:56<34:37,  1.66s/it, loss=0.1756, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1605/2855 [1:06:56<33:27,  1.61s/it, loss=0.1756, lr=2.74e-05]
Epoch 2/8:  56%|█████▌    | 1605/2855 [1:06:58<33:27,  1.61s/it, loss=0.0833, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1606/2855 [1:06:58<34:01,  1.63s/it, loss=0.0833, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1606/2855 [1:06:59<34:01,  1.63s/it, loss=0.0356, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1607/2855 [1:06:59<34:08,  1.64s/it, loss=0.0356, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1607/2855 [1:07:01<34:08,  1.64s/it, loss=0.2207, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1608/2855 [1:07:01<33:42,  1.62s/it, loss=0.2207, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1608/2855 [1:07:03<33:42,  1.62s/it, loss=0.1225, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1609/2855 [1:07:03<34:08,  1.64s/it, loss=0.1225, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1609/2855 [1:07:04<34:08,  1.64s/it, loss=0.1599, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1610/2855 [1:07:04<34:25,  1.66s/it, loss=0.1599, lr=2.74e-05]
Epoch 2/8:  56%|█████▋    | 1610/2855 [1:07:06<34:25,  1.66s/it, loss=0.2741, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1611/2855 [1:07:06<34:14,  1.65s/it, loss=0.2741, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1611/2855 [1:07:08<34:14,  1.65s/it, loss=0.0971, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1612/2855 [1:07:08<36:41,  1.77s/it, loss=0.0971, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1612/2855 [1:07:10<36:41,  1.77s/it, loss=0.1187, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1613/2855 [1:07:10<36:19,  1.75s/it, loss=0.1187, lr=2.73e-05]
Epoch 2/8:  56%|█████▋    | 1613/2855 [1:07:11<36:19,  1.75s/it, loss=0.1492, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1614/2855 [1:07:11<35:18,  1.71s/it, loss=0.1492, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1614/2855 [1:07:13<35:18,  1.71s/it, loss=0.4098, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1615/2855 [1:07:13<35:32,  1.72s/it, loss=0.4098, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1615/2855 [1:07:15<35:32,  1.72s/it, loss=0.2744, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1616/2855 [1:07:15<35:29,  1.72s/it, loss=0.2744, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1616/2855 [1:07:16<35:29,  1.72s/it, loss=0.1884, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1617/2855 [1:07:16<35:03,  1.70s/it, loss=0.1884, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1617/2855 [1:07:18<35:03,  1.70s/it, loss=0.1049, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1618/2855 [1:07:18<35:22,  1.72s/it, loss=0.1049, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1618/2855 [1:07:20<35:22,  1.72s/it, loss=0.1188, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1619/2855 [1:07:20<34:22,  1.67s/it, loss=0.1188, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1619/2855 [1:07:21<34:22,  1.67s/it, loss=0.1070, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1620/2855 [1:07:21<32:22,  1.57s/it, loss=0.1070, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1620/2855 [1:07:22<32:22,  1.57s/it, loss=0.1566, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1621/2855 [1:07:22<31:25,  1.53s/it, loss=0.1566, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1621/2855 [1:07:24<31:25,  1.53s/it, loss=0.1880, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1622/2855 [1:07:24<31:02,  1.51s/it, loss=0.1880, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1622/2855 [1:07:26<31:02,  1.51s/it, loss=0.1422, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1623/2855 [1:07:26<31:29,  1.53s/it, loss=0.1422, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1623/2855 [1:07:27<31:29,  1.53s/it, loss=0.1353, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1624/2855 [1:07:27<32:47,  1.60s/it, loss=0.1353, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1624/2855 [1:07:29<32:47,  1.60s/it, loss=0.0038, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1625/2855 [1:07:29<33:57,  1.66s/it, loss=0.0038, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1625/2855 [1:07:31<33:57,  1.66s/it, loss=0.0660, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1626/2855 [1:07:31<33:17,  1.63s/it, loss=0.0660, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1626/2855 [1:07:32<33:17,  1.63s/it, loss=0.1542, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1627/2855 [1:07:32<33:39,  1.64s/it, loss=0.1542, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1627/2855 [1:07:34<33:39,  1.64s/it, loss=0.1014, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1628/2855 [1:07:34<34:29,  1.69s/it, loss=0.1014, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1628/2855 [1:07:36<34:29,  1.69s/it, loss=0.3005, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1629/2855 [1:07:36<33:39,  1.65s/it, loss=0.3005, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1629/2855 [1:07:37<33:39,  1.65s/it, loss=0.2227, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1630/2855 [1:07:37<34:08,  1.67s/it, loss=0.2227, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1630/2855 [1:07:39<34:08,  1.67s/it, loss=0.0940, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1631/2855 [1:07:39<34:31,  1.69s/it, loss=0.0940, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1631/2855 [1:07:41<34:31,  1.69s/it, loss=0.2727, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1632/2855 [1:07:41<33:31,  1.65s/it, loss=0.2727, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1632/2855 [1:07:42<33:31,  1.65s/it, loss=0.1813, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1633/2855 [1:07:42<33:03,  1.62s/it, loss=0.1813, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1633/2855 [1:07:44<33:03,  1.62s/it, loss=0.0539, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1634/2855 [1:07:44<33:41,  1.66s/it, loss=0.0539, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1634/2855 [1:07:46<33:41,  1.66s/it, loss=0.1312, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1635/2855 [1:07:46<33:25,  1.64s/it, loss=0.1312, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1635/2855 [1:07:47<33:25,  1.64s/it, loss=0.0832, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1636/2855 [1:07:47<32:33,  1.60s/it, loss=0.0832, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1636/2855 [1:07:49<32:33,  1.60s/it, loss=0.1171, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1637/2855 [1:07:49<31:48,  1.57s/it, loss=0.1171, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1637/2855 [1:07:50<31:48,  1.57s/it, loss=0.2593, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1638/2855 [1:07:50<31:50,  1.57s/it, loss=0.2593, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1638/2855 [1:07:52<31:50,  1.57s/it, loss=0.2324, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1639/2855 [1:07:52<32:03,  1.58s/it, loss=0.2324, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1639/2855 [1:07:54<32:03,  1.58s/it, loss=0.1046, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1640/2855 [1:07:54<33:26,  1.65s/it, loss=0.1046, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1640/2855 [1:07:55<33:26,  1.65s/it, loss=0.0573, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1641/2855 [1:07:55<33:55,  1.68s/it, loss=0.0573, lr=2.73e-05]
Epoch 2/8:  57%|█████▋    | 1641/2855 [1:07:57<33:55,  1.68s/it, loss=0.1096, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1642/2855 [1:07:57<33:03,  1.64s/it, loss=0.1096, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1642/2855 [1:07:58<33:03,  1.64s/it, loss=0.2754, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1643/2855 [1:07:58<32:32,  1.61s/it, loss=0.2754, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1643/2855 [1:08:00<32:32,  1.61s/it, loss=0.0778, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1644/2855 [1:08:00<34:22,  1.70s/it, loss=0.0778, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1644/2855 [1:08:02<34:22,  1.70s/it, loss=0.2680, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1645/2855 [1:08:02<35:22,  1.75s/it, loss=0.2680, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1645/2855 [1:08:04<35:22,  1.75s/it, loss=0.0529, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1646/2855 [1:08:04<36:22,  1.80s/it, loss=0.0529, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1646/2855 [1:08:06<36:22,  1.80s/it, loss=0.2025, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1647/2855 [1:08:06<39:17,  1.95s/it, loss=0.2025, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1647/2855 [1:08:08<39:17,  1.95s/it, loss=0.1404, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1648/2855 [1:08:08<40:10,  2.00s/it, loss=0.1404, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1648/2855 [1:08:11<40:10,  2.00s/it, loss=0.1723, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1649/2855 [1:08:11<40:31,  2.02s/it, loss=0.1723, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1649/2855 [1:08:13<40:31,  2.02s/it, loss=0.1433, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1650/2855 [1:08:13<41:31,  2.07s/it, loss=0.1433, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1650/2855 [1:08:15<41:31,  2.07s/it, loss=0.2487, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1651/2855 [1:08:15<40:41,  2.03s/it, loss=0.2487, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1651/2855 [1:08:20<40:41,  2.03s/it, loss=0.2226, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1652/2855 [1:08:20<1:02:12,  3.10s/it, loss=0.2226, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1652/2855 [1:08:22<1:02:12,  3.10s/it, loss=0.0470, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1653/2855 [1:08:22<52:09,  2.60s/it, loss=0.0470, lr=2.73e-05]  
Epoch 2/8:  58%|█████▊    | 1653/2855 [1:08:24<52:09,  2.60s/it, loss=0.1306, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1654/2855 [1:08:24<48:59,  2.45s/it, loss=0.1306, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1654/2855 [1:08:26<48:59,  2.45s/it, loss=0.1351, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1655/2855 [1:08:26<45:13,  2.26s/it, loss=0.1351, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1655/2855 [1:08:27<45:13,  2.26s/it, loss=0.1269, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1656/2855 [1:08:27<41:55,  2.10s/it, loss=0.1269, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1656/2855 [1:08:29<41:55,  2.10s/it, loss=0.3570, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1657/2855 [1:08:29<40:43,  2.04s/it, loss=0.3570, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1657/2855 [1:08:31<40:43,  2.04s/it, loss=0.1008, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1658/2855 [1:08:31<41:12,  2.07s/it, loss=0.1008, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1658/2855 [1:08:33<41:12,  2.07s/it, loss=0.1166, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1659/2855 [1:08:33<40:28,  2.03s/it, loss=0.1166, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1659/2855 [1:08:35<40:28,  2.03s/it, loss=0.1053, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1660/2855 [1:08:35<39:43,  1.99s/it, loss=0.1053, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1660/2855 [1:08:37<39:43,  1.99s/it, loss=0.1231, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1661/2855 [1:08:37<38:44,  1.95s/it, loss=0.1231, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1661/2855 [1:08:39<38:44,  1.95s/it, loss=0.1368, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1662/2855 [1:08:39<38:44,  1.95s/it, loss=0.1368, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1662/2855 [1:08:41<38:44,  1.95s/it, loss=0.2355, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1663/2855 [1:08:41<38:04,  1.92s/it, loss=0.2355, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1663/2855 [1:08:43<38:04,  1.92s/it, loss=0.1618, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1664/2855 [1:08:43<38:05,  1.92s/it, loss=0.1618, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1664/2855 [1:08:45<38:05,  1.92s/it, loss=0.2015, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1665/2855 [1:08:45<38:25,  1.94s/it, loss=0.2015, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1665/2855 [1:08:47<38:25,  1.94s/it, loss=0.2002, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1666/2855 [1:08:47<38:19,  1.93s/it, loss=0.2002, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1666/2855 [1:08:49<38:19,  1.93s/it, loss=0.1484, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1667/2855 [1:08:49<37:39,  1.90s/it, loss=0.1484, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1667/2855 [1:08:50<37:39,  1.90s/it, loss=0.1178, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1668/2855 [1:08:50<37:32,  1.90s/it, loss=0.1178, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1668/2855 [1:08:52<37:32,  1.90s/it, loss=0.0527, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1669/2855 [1:08:52<37:53,  1.92s/it, loss=0.0527, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1669/2855 [1:08:54<37:53,  1.92s/it, loss=0.2199, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1670/2855 [1:08:54<38:01,  1.93s/it, loss=0.2199, lr=2.73e-05]
Epoch 2/8:  58%|█████▊    | 1670/2855 [1:08:56<38:01,  1.93s/it, loss=0.1372, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1671/2855 [1:08:56<37:40,  1.91s/it, loss=0.1372, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1671/2855 [1:08:58<37:40,  1.91s/it, loss=0.2210, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1672/2855 [1:08:58<39:30,  2.00s/it, loss=0.2210, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1672/2855 [1:09:00<39:30,  2.00s/it, loss=0.1203, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1673/2855 [1:09:00<38:43,  1.97s/it, loss=0.1203, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1673/2855 [1:09:02<38:43,  1.97s/it, loss=0.2048, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1674/2855 [1:09:02<38:24,  1.95s/it, loss=0.2048, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1674/2855 [1:09:04<38:24,  1.95s/it, loss=0.3945, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1675/2855 [1:09:04<38:22,  1.95s/it, loss=0.3945, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1675/2855 [1:09:06<38:22,  1.95s/it, loss=0.1781, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1676/2855 [1:09:06<38:29,  1.96s/it, loss=0.1781, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1676/2855 [1:09:08<38:29,  1.96s/it, loss=0.0988, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1677/2855 [1:09:08<38:55,  1.98s/it, loss=0.0988, lr=2.73e-05]
Epoch 2/8:  59%|█████▊    | 1677/2855 [1:09:10<38:55,  1.98s/it, loss=0.1985, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1678/2855 [1:09:10<37:48,  1.93s/it, loss=0.1985, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1678/2855 [1:09:12<37:48,  1.93s/it, loss=0.2826, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1679/2855 [1:09:12<37:16,  1.90s/it, loss=0.2826, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1679/2855 [1:09:14<37:16,  1.90s/it, loss=0.1130, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1680/2855 [1:09:14<38:36,  1.97s/it, loss=0.1130, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1680/2855 [1:09:16<38:36,  1.97s/it, loss=0.1642, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1681/2855 [1:09:16<38:43,  1.98s/it, loss=0.1642, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1681/2855 [1:09:18<38:43,  1.98s/it, loss=0.1259, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1682/2855 [1:09:18<38:54,  1.99s/it, loss=0.1259, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1682/2855 [1:09:20<38:54,  1.99s/it, loss=0.2251, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1683/2855 [1:09:20<38:18,  1.96s/it, loss=0.2251, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1683/2855 [1:09:22<38:18,  1.96s/it, loss=0.2268, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1684/2855 [1:09:22<37:48,  1.94s/it, loss=0.2268, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1684/2855 [1:09:24<37:48,  1.94s/it, loss=0.1649, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1685/2855 [1:09:24<38:29,  1.97s/it, loss=0.1649, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1685/2855 [1:09:26<38:29,  1.97s/it, loss=0.3184, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1686/2855 [1:09:26<38:05,  1.95s/it, loss=0.3184, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1686/2855 [1:09:28<38:05,  1.95s/it, loss=0.2933, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1687/2855 [1:09:28<37:41,  1.94s/it, loss=0.2933, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1687/2855 [1:09:30<37:41,  1.94s/it, loss=0.2034, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1688/2855 [1:09:30<37:41,  1.94s/it, loss=0.2034, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1688/2855 [1:09:32<37:41,  1.94s/it, loss=0.0455, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1689/2855 [1:09:32<38:07,  1.96s/it, loss=0.0455, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1689/2855 [1:09:34<38:07,  1.96s/it, loss=0.0781, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1690/2855 [1:09:34<38:32,  1.98s/it, loss=0.0781, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1690/2855 [1:09:36<38:32,  1.98s/it, loss=0.1193, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1691/2855 [1:09:36<40:58,  2.11s/it, loss=0.1193, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1691/2855 [1:09:38<40:58,  2.11s/it, loss=0.2317, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1692/2855 [1:09:38<38:24,  1.98s/it, loss=0.2317, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1692/2855 [1:09:40<38:24,  1.98s/it, loss=0.1380, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1693/2855 [1:09:40<38:17,  1.98s/it, loss=0.1380, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1693/2855 [1:09:42<38:17,  1.98s/it, loss=0.1152, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1694/2855 [1:09:42<37:55,  1.96s/it, loss=0.1152, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1694/2855 [1:09:44<37:55,  1.96s/it, loss=0.2374, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1695/2855 [1:09:44<37:43,  1.95s/it, loss=0.2374, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1695/2855 [1:09:45<37:43,  1.95s/it, loss=0.1275, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1696/2855 [1:09:45<37:01,  1.92s/it, loss=0.1275, lr=2.73e-05]
Epoch 2/8:  59%|█████▉    | 1696/2855 [1:09:47<37:01,  1.92s/it, loss=0.0966, lr=2.72e-05]
Epoch 2/8:  59%|█████▉    | 1697/2855 [1:09:47<36:55,  1.91s/it, loss=0.0966, lr=2.72e-05]
Epoch 2/8:  59%|█████▉    | 1697/2855 [1:09:49<36:55,  1.91s/it, loss=0.1667, lr=2.72e-05]
Epoch 2/8:  59%|█████▉    | 1698/2855 [1:09:49<37:41,  1.95s/it, loss=0.1667, lr=2.72e-05]
Epoch 2/8:  59%|█████▉    | 1698/2855 [1:09:51<37:41,  1.95s/it, loss=0.1567, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1699/2855 [1:09:51<37:18,  1.94s/it, loss=0.1567, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1699/2855 [1:09:53<37:18,  1.94s/it, loss=0.1487, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1700/2855 [1:09:53<38:08,  1.98s/it, loss=0.1487, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1700/2855 [1:09:55<38:08,  1.98s/it, loss=0.2544, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1701/2855 [1:09:55<38:01,  1.98s/it, loss=0.2544, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1701/2855 [1:09:57<38:01,  1.98s/it, loss=0.2141, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1702/2855 [1:09:57<37:54,  1.97s/it, loss=0.2141, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1702/2855 [1:09:59<37:54,  1.97s/it, loss=0.2453, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1703/2855 [1:09:59<38:15,  1.99s/it, loss=0.2453, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1703/2855 [1:10:01<38:15,  1.99s/it, loss=0.4119, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1704/2855 [1:10:01<38:56,  2.03s/it, loss=0.4119, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1704/2855 [1:10:04<38:56,  2.03s/it, loss=0.1637, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1705/2855 [1:10:04<39:53,  2.08s/it, loss=0.1637, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1705/2855 [1:10:05<39:53,  2.08s/it, loss=0.0667, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1706/2855 [1:10:05<38:31,  2.01s/it, loss=0.0667, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1706/2855 [1:10:07<38:31,  2.01s/it, loss=0.1243, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1707/2855 [1:10:07<38:03,  1.99s/it, loss=0.1243, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1707/2855 [1:10:09<38:03,  1.99s/it, loss=0.0005, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1708/2855 [1:10:09<38:21,  2.01s/it, loss=0.0005, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1708/2855 [1:10:11<38:21,  2.01s/it, loss=0.0342, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1709/2855 [1:10:11<38:05,  1.99s/it, loss=0.0342, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1709/2855 [1:10:14<38:05,  1.99s/it, loss=0.0635, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1710/2855 [1:10:14<39:00,  2.04s/it, loss=0.0635, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1710/2855 [1:10:15<39:00,  2.04s/it, loss=0.2290, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1711/2855 [1:10:15<37:54,  1.99s/it, loss=0.2290, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1711/2855 [1:10:17<37:54,  1.99s/it, loss=0.2284, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1712/2855 [1:10:17<37:20,  1.96s/it, loss=0.2284, lr=2.72e-05]
Epoch 2/8:  60%|█████▉    | 1712/2855 [1:10:19<37:20,  1.96s/it, loss=0.0636, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1713/2855 [1:10:19<38:05,  2.00s/it, loss=0.0636, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1713/2855 [1:10:21<38:05,  2.00s/it, loss=0.1516, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1714/2855 [1:10:21<37:48,  1.99s/it, loss=0.1516, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1714/2855 [1:10:24<37:48,  1.99s/it, loss=0.1342, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1715/2855 [1:10:24<38:49,  2.04s/it, loss=0.1342, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1715/2855 [1:10:25<38:49,  2.04s/it, loss=0.1476, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1716/2855 [1:10:25<38:02,  2.00s/it, loss=0.1476, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1716/2855 [1:10:27<38:02,  2.00s/it, loss=0.2045, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1717/2855 [1:10:27<37:38,  1.98s/it, loss=0.2045, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1717/2855 [1:10:29<37:38,  1.98s/it, loss=0.0908, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1718/2855 [1:10:29<38:16,  2.02s/it, loss=0.0908, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1718/2855 [1:10:31<38:16,  2.02s/it, loss=0.2635, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1719/2855 [1:10:31<37:53,  2.00s/it, loss=0.2635, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1719/2855 [1:10:34<37:53,  2.00s/it, loss=0.2231, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1720/2855 [1:10:34<38:37,  2.04s/it, loss=0.2231, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1720/2855 [1:10:35<38:37,  2.04s/it, loss=0.1588, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1721/2855 [1:10:35<37:33,  1.99s/it, loss=0.1588, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1721/2855 [1:10:37<37:33,  1.99s/it, loss=0.1730, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1722/2855 [1:10:37<37:15,  1.97s/it, loss=0.1730, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1722/2855 [1:10:40<37:15,  1.97s/it, loss=0.1694, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1723/2855 [1:10:40<39:24,  2.09s/it, loss=0.1694, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1723/2855 [1:10:42<39:24,  2.09s/it, loss=0.2412, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1724/2855 [1:10:42<39:55,  2.12s/it, loss=0.2412, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1724/2855 [1:10:44<39:55,  2.12s/it, loss=0.0783, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1725/2855 [1:10:44<41:14,  2.19s/it, loss=0.0783, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1725/2855 [1:10:47<41:14,  2.19s/it, loss=0.1775, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1726/2855 [1:10:47<41:54,  2.23s/it, loss=0.1775, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1726/2855 [1:10:49<41:54,  2.23s/it, loss=0.1062, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1727/2855 [1:10:49<41:20,  2.20s/it, loss=0.1062, lr=2.72e-05]
Epoch 2/8:  60%|██████    | 1727/2855 [1:10:51<41:20,  2.20s/it, loss=0.0897, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1728/2855 [1:10:51<40:02,  2.13s/it, loss=0.0897, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1728/2855 [1:10:53<40:02,  2.13s/it, loss=0.1561, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1729/2855 [1:10:53<39:03,  2.08s/it, loss=0.1561, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1729/2855 [1:10:55<39:03,  2.08s/it, loss=0.1702, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1730/2855 [1:10:55<39:52,  2.13s/it, loss=0.1702, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1730/2855 [1:10:59<39:52,  2.13s/it, loss=0.1342, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1731/2855 [1:10:59<49:09,  2.62s/it, loss=0.1342, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1731/2855 [1:11:01<49:09,  2.62s/it, loss=0.0971, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1732/2855 [1:11:01<45:18,  2.42s/it, loss=0.0971, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1732/2855 [1:11:03<45:18,  2.42s/it, loss=0.1501, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1733/2855 [1:11:03<43:22,  2.32s/it, loss=0.1501, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1733/2855 [1:11:05<43:22,  2.32s/it, loss=0.1226, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1734/2855 [1:11:05<43:11,  2.31s/it, loss=0.1226, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1734/2855 [1:11:07<43:11,  2.31s/it, loss=0.0014, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1735/2855 [1:11:07<41:55,  2.25s/it, loss=0.0014, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1735/2855 [1:11:09<41:55,  2.25s/it, loss=0.1001, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1736/2855 [1:11:09<41:50,  2.24s/it, loss=0.1001, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1736/2855 [1:11:11<41:50,  2.24s/it, loss=0.0499, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1737/2855 [1:11:11<39:45,  2.13s/it, loss=0.0499, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1737/2855 [1:11:13<39:45,  2.13s/it, loss=0.1030, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1738/2855 [1:11:13<39:41,  2.13s/it, loss=0.1030, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1738/2855 [1:11:15<39:41,  2.13s/it, loss=0.0046, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1739/2855 [1:11:15<39:26,  2.12s/it, loss=0.0046, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1739/2855 [1:11:17<39:26,  2.12s/it, loss=0.0859, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1740/2855 [1:11:17<38:10,  2.05s/it, loss=0.0859, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1740/2855 [1:11:20<38:10,  2.05s/it, loss=0.2039, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1741/2855 [1:11:20<39:58,  2.15s/it, loss=0.2039, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1741/2855 [1:11:22<39:58,  2.15s/it, loss=0.1336, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1742/2855 [1:11:22<39:07,  2.11s/it, loss=0.1336, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1742/2855 [1:11:24<39:07,  2.11s/it, loss=0.0804, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1743/2855 [1:11:24<40:04,  2.16s/it, loss=0.0804, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1743/2855 [1:11:26<40:04,  2.16s/it, loss=0.1264, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1744/2855 [1:11:26<40:50,  2.21s/it, loss=0.1264, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1744/2855 [1:11:29<40:50,  2.21s/it, loss=0.0966, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1745/2855 [1:11:29<40:54,  2.21s/it, loss=0.0966, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1745/2855 [1:11:31<40:54,  2.21s/it, loss=0.1502, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1746/2855 [1:11:31<41:47,  2.26s/it, loss=0.1502, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1746/2855 [1:11:33<41:47,  2.26s/it, loss=0.1394, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1747/2855 [1:11:33<40:14,  2.18s/it, loss=0.1394, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1747/2855 [1:11:35<40:14,  2.18s/it, loss=0.0658, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1748/2855 [1:11:35<41:20,  2.24s/it, loss=0.0658, lr=2.72e-05]
Epoch 2/8:  61%|██████    | 1748/2855 [1:11:37<41:20,  2.24s/it, loss=0.1565, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1749/2855 [1:11:37<40:22,  2.19s/it, loss=0.1565, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1749/2855 [1:11:40<40:22,  2.19s/it, loss=0.1854, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1750/2855 [1:11:40<40:32,  2.20s/it, loss=0.1854, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1750/2855 [1:11:42<40:32,  2.20s/it, loss=0.1423, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1751/2855 [1:11:42<39:05,  2.12s/it, loss=0.1423, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1751/2855 [1:11:43<39:05,  2.12s/it, loss=0.1584, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1752/2855 [1:11:43<38:07,  2.07s/it, loss=0.1584, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1752/2855 [1:11:46<38:07,  2.07s/it, loss=0.1513, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1753/2855 [1:11:46<38:40,  2.11s/it, loss=0.1513, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1753/2855 [1:11:48<38:40,  2.11s/it, loss=0.0711, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1754/2855 [1:11:48<38:28,  2.10s/it, loss=0.0711, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1754/2855 [1:11:50<38:28,  2.10s/it, loss=0.0496, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1755/2855 [1:11:50<40:26,  2.21s/it, loss=0.0496, lr=2.72e-05]
Epoch 2/8:  61%|██████▏   | 1755/2855 [1:11:52<40:26,  2.21s/it, loss=0.2826, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1756/2855 [1:11:52<39:00,  2.13s/it, loss=0.2826, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1756/2855 [1:11:54<39:00,  2.13s/it, loss=0.0392, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1757/2855 [1:11:54<39:58,  2.18s/it, loss=0.0392, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1757/2855 [1:11:56<39:58,  2.18s/it, loss=0.1953, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1758/2855 [1:11:56<39:05,  2.14s/it, loss=0.1953, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1758/2855 [1:11:58<39:05,  2.14s/it, loss=0.2825, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1759/2855 [1:11:58<38:14,  2.09s/it, loss=0.2825, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1759/2855 [1:12:01<38:14,  2.09s/it, loss=0.0813, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1760/2855 [1:12:01<38:50,  2.13s/it, loss=0.0813, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1760/2855 [1:12:03<38:50,  2.13s/it, loss=0.1527, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1761/2855 [1:12:03<38:01,  2.09s/it, loss=0.1527, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1761/2855 [1:12:05<38:01,  2.09s/it, loss=0.1646, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1762/2855 [1:12:05<38:35,  2.12s/it, loss=0.1646, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1762/2855 [1:12:07<38:35,  2.12s/it, loss=0.2191, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1763/2855 [1:12:07<37:43,  2.07s/it, loss=0.2191, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1763/2855 [1:12:09<37:43,  2.07s/it, loss=0.0625, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1764/2855 [1:12:09<38:57,  2.14s/it, loss=0.0625, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1764/2855 [1:12:12<38:57,  2.14s/it, loss=0.2212, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1765/2855 [1:12:12<40:48,  2.25s/it, loss=0.2212, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1765/2855 [1:12:14<40:48,  2.25s/it, loss=0.1022, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1766/2855 [1:12:14<39:12,  2.16s/it, loss=0.1022, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1766/2855 [1:12:16<39:12,  2.16s/it, loss=0.1099, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1767/2855 [1:12:16<40:06,  2.21s/it, loss=0.1099, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1767/2855 [1:12:18<40:06,  2.21s/it, loss=0.1198, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1768/2855 [1:12:18<41:23,  2.29s/it, loss=0.1198, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1768/2855 [1:12:21<41:23,  2.29s/it, loss=0.1153, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1769/2855 [1:12:21<43:37,  2.41s/it, loss=0.1153, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1769/2855 [1:12:24<43:37,  2.41s/it, loss=0.1043, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1770/2855 [1:12:24<48:15,  2.67s/it, loss=0.1043, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1770/2855 [1:12:26<48:15,  2.67s/it, loss=0.1977, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1771/2855 [1:12:26<44:06,  2.44s/it, loss=0.1977, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1771/2855 [1:12:29<44:06,  2.44s/it, loss=0.1602, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1772/2855 [1:12:29<43:23,  2.40s/it, loss=0.1602, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1772/2855 [1:12:31<43:23,  2.40s/it, loss=0.0497, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1773/2855 [1:12:31<44:31,  2.47s/it, loss=0.0497, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1773/2855 [1:12:33<44:31,  2.47s/it, loss=0.0734, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1774/2855 [1:12:33<42:23,  2.35s/it, loss=0.0734, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1774/2855 [1:12:36<42:23,  2.35s/it, loss=0.1217, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1775/2855 [1:12:36<41:46,  2.32s/it, loss=0.1217, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1775/2855 [1:12:38<41:46,  2.32s/it, loss=0.2299, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1776/2855 [1:12:38<41:50,  2.33s/it, loss=0.2299, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1776/2855 [1:12:40<41:50,  2.33s/it, loss=0.1879, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1777/2855 [1:12:40<40:24,  2.25s/it, loss=0.1879, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1777/2855 [1:12:42<40:24,  2.25s/it, loss=0.1654, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1778/2855 [1:12:42<39:39,  2.21s/it, loss=0.1654, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1778/2855 [1:12:44<39:39,  2.21s/it, loss=0.0709, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1779/2855 [1:12:44<38:47,  2.16s/it, loss=0.0709, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1779/2855 [1:12:46<38:47,  2.16s/it, loss=0.2322, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1780/2855 [1:12:46<40:01,  2.23s/it, loss=0.2322, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1780/2855 [1:12:49<40:01,  2.23s/it, loss=0.1555, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1781/2855 [1:12:49<40:30,  2.26s/it, loss=0.1555, lr=2.72e-05]
Epoch 2/8:  62%|██████▏   | 1781/2855 [1:12:51<40:30,  2.26s/it, loss=0.0419, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1782/2855 [1:12:51<40:54,  2.29s/it, loss=0.0419, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1782/2855 [1:12:53<40:54,  2.29s/it, loss=0.1410, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1783/2855 [1:12:53<39:27,  2.21s/it, loss=0.1410, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1783/2855 [1:12:56<39:27,  2.21s/it, loss=0.1259, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1784/2855 [1:12:56<39:57,  2.24s/it, loss=0.1259, lr=2.71e-05]
Epoch 2/8:  62%|██████▏   | 1784/2855 [1:12:58<39:57,  2.24s/it, loss=0.0879, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1785/2855 [1:12:58<40:45,  2.29s/it, loss=0.0879, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1785/2855 [1:13:00<40:45,  2.29s/it, loss=0.0646, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1786/2855 [1:13:00<40:44,  2.29s/it, loss=0.0646, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1786/2855 [1:13:03<40:44,  2.29s/it, loss=0.2418, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1787/2855 [1:13:03<41:38,  2.34s/it, loss=0.2418, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1787/2855 [1:13:05<41:38,  2.34s/it, loss=0.0338, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1788/2855 [1:13:05<40:31,  2.28s/it, loss=0.0338, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1788/2855 [1:13:07<40:31,  2.28s/it, loss=0.2783, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1789/2855 [1:13:07<40:08,  2.26s/it, loss=0.2783, lr=2.71e-05]
Epoch 2/8:  63%|██████���   | 1789/2855 [1:13:09<40:08,  2.26s/it, loss=0.1327, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1790/2855 [1:13:09<39:07,  2.20s/it, loss=0.1327, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1790/2855 [1:13:11<39:07,  2.20s/it, loss=0.2220, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1791/2855 [1:13:11<39:38,  2.24s/it, loss=0.2220, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1791/2855 [1:13:14<39:38,  2.24s/it, loss=0.1327, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1792/2855 [1:13:14<42:44,  2.41s/it, loss=0.1327, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1792/2855 [1:13:17<42:44,  2.41s/it, loss=0.1272, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1793/2855 [1:13:17<44:28,  2.51s/it, loss=0.1272, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1793/2855 [1:13:29<44:28,  2.51s/it, loss=0.0755, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1794/2855 [1:13:29<1:34:20,  5.34s/it, loss=0.0755, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1794/2855 [1:13:30<1:34:20,  5.34s/it, loss=0.2501, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1795/2855 [1:13:30<1:13:47,  4.18s/it, loss=0.2501, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1795/2855 [1:13:43<1:13:47,  4.18s/it, loss=0.3393, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1796/2855 [1:13:43<1:58:59,  6.74s/it, loss=0.3393, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1796/2855 [1:14:03<1:58:59,  6.74s/it, loss=0.1426, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1797/2855 [1:14:03<3:10:15, 10.79s/it, loss=0.1426, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1797/2855 [2:42:23<3:10:15, 10.79s/it, loss=0.0975, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1798/2855 [2:42:23<469:00:30, 1597.38s/it, loss=0.0975, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1798/2855 [2:42:29<469:00:30, 1597.38s/it, loss=0.0095, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1799/2855 [2:42:29<328:33:11, 1120.07s/it, loss=0.0095, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1799/2855 [4:15:35<328:33:11, 1120.07s/it, loss=0.1382, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1800/2855 [4:15:35<720:51:00, 2459.77s/it, loss=0.1382, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1800/2855 [5:41:20<720:51:00, 2459.77s/it, loss=0.0972, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1801/2855 [5:41:20<956:00:35, 3265.31s/it, loss=0.0972, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1801/2855 [5:41:25<956:00:35, 3265.31s/it, loss=0.2867, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1802/2855 [5:41:25<669:00:11, 2287.19s/it, loss=0.2867, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1802/2855 [6:17:23<669:00:11, 2287.19s/it, loss=0.1633, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1803/2855 [6:17:23<657:06:17, 2248.65s/it, loss=0.1633, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1803/2855 [6:17:31<657:06:17, 2248.65s/it, loss=0.1567, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1804/2855 [6:17:31<460:14:46, 1576.49s/it, loss=0.1567, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1804/2855 [6:17:57<460:14:46, 1576.49s/it, loss=0.1132, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1805/2855 [6:17:57<324:06:34, 1111.23s/it, loss=0.1132, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1805/2855 [7:18:20<324:06:34, 1111.23s/it, loss=0.1185, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1806/2855 [7:18:20<543:23:49, 1864.85s/it, loss=0.1185, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1806/2855 [7:18:24<543:23:49, 1864.85s/it, loss=0.1451, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1807/2855 [7:18:24<380:17:16, 1306.33s/it, loss=0.1451, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1807/2855 [7:19:04<380:17:16, 1306.33s/it, loss=0.1643, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1808/2855 [7:19:04<269:31:05, 926.71s/it, loss=0.1643, lr=2.71e-05] 
Epoch 2/8:  63%|██████▎   | 1808/2855 [7:21:49<269:31:05, 926.71s/it, loss=0.2034, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1809/2855 [7:21:49<202:50:18, 698.11s/it, loss=0.2034, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1809/2855 [7:22:25<202:50:18, 698.11s/it, loss=0.1063, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1810/2855 [7:22:25<144:59:47, 499.51s/it, loss=0.1063, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1810/2855 [7:31:39<144:59:47, 499.51s/it, loss=0.2075, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1811/2855 [7:31:39<149:34:12, 515.76s/it, loss=0.2075, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1811/2855 [7:32:38<149:34:12, 515.76s/it, loss=0.2285, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1812/2855 [7:32:38<109:44:52, 378.80s/it, loss=0.2285, lr=2.71e-05]
Epoch 2/8:  63%|██████▎   | 1812/2855 [7:32:48<109:44:52, 378.80s/it, loss=0.1764, lr=2.71e-05]
Epoch 2/8:  64%|█████��▎   | 1813/2855 [7:32:48<77:36:34, 268.13s/it, loss=0.1764, lr=2.71e-05] 
Epoch 2/8:  64%|██████▎   | 1813/2855 [7:33:01<77:36:34, 268.13s/it, loss=0.1937, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1814/2855 [7:33:01<55:21:52, 191.46s/it, loss=0.1937, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1814/2855 [7:33:14<55:21:52, 191.46s/it, loss=0.0952, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1815/2855 [7:33:14<39:52:18, 138.02s/it, loss=0.0952, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1815/2855 [7:33:47<39:52:18, 138.02s/it, loss=0.1912, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1816/2855 [7:33:47<30:45:43, 106.59s/it, loss=0.1912, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1816/2855 [7:33:55<30:45:43, 106.59s/it, loss=0.1173, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1817/2855 [7:33:55<22:12:16, 77.01s/it, loss=0.1173, lr=2.71e-05] 
Epoch 2/8:  64%|██████▎   | 1817/2855 [7:34:22<22:12:16, 77.01s/it, loss=0.1381, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1818/2855 [7:34:22<17:51:25, 61.99s/it, loss=0.1381, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1818/2855 [7:34:30<17:51:25, 61.99s/it, loss=0.1075, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1819/2855 [7:34:30<13:09:34, 45.73s/it, loss=0.1075, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1819/2855 [8:19:23<13:09:34, 45.73s/it, loss=0.1144, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1820/2855 [8:19:23<241:28:26, 839.91s/it, loss=0.1144, lr=2.71e-05]
Epoch 2/8:  64%|██████▎   | 1820/2855 [8:19:26<241:28:26, 839.91s/it, loss=0.0659, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1821/2855 [8:19:26<169:09:42, 588.96s/it, loss=0.0659, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1821/2855 [8:19:38<169:09:42, 588.96s/it, loss=0.2355, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1822/2855 [8:19:38<119:20:04, 415.88s/it, loss=0.2355, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1822/2855 [8:20:01<119:20:04, 415.88s/it, loss=0.1292, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1823/2855 [8:20:01<85:25:34, 298.00s/it, loss=0.1292, lr=2.71e-05] 
Epoch 2/8:  64%|██████▍   | 1823/2855 [8:58:00<85:25:34, 298.00s/it, loss=0.1777, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1824/2855 [8:58:00<255:30:18, 892.16s/it, loss=0.1777, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1824/2855 [8:58:03<255:30:18, 892.16s/it, loss=0.1325, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1825/2855 [8:58:03<178:54:50, 625.33s/it, loss=0.1325, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1825/2855 [8:58:12<178:54:50, 625.33s/it, loss=0.0292, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1826/2855 [8:58:12<125:58:02, 440.70s/it, loss=0.0292, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1826/2855 [8:58:32<125:58:02, 440.70s/it, loss=0.1974, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1827/2855 [8:58:32<89:44:32, 314.27s/it, loss=0.1974, lr=2.71e-05] 
Epoch 2/8:  64%|██████▍   | 1827/2855 [8:59:13<89:44:32, 314.27s/it, loss=0.0643, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1828/2855 [8:59:13<66:16:11, 232.30s/it, loss=0.0643, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1828/2855 [8:59:16<66:16:11, 232.30s/it, loss=0.1799, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1829/2855 [8:59:16<46:39:30, 163.71s/it, loss=0.1799, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1829/2855 [8:59:30<46:39:30, 163.71s/it, loss=0.1097, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1830/2855 [8:59:30<33:45:35, 118.57s/it, loss=0.1097, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1830/2855 [8:59:42<33:45:35, 118.57s/it, loss=0.1363, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1831/2855 [8:59:42<24:40:14, 86.73s/it, loss=0.1363, lr=2.71e-05] 
Epoch 2/8:  64%|██████▍   | 1831/2855 [9:09:26<24:40:14, 86.73s/it, loss=0.0062, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1832/2855 [9:09:26<67:03:45, 236.00s/it, loss=0.0062, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1832/2855 [9:09:37<67:03:45, 236.00s/it, loss=0.1149, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1833/2855 [9:09:37<47:46:10, 168.27s/it, loss=0.1149, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1833/2855 [9:09:50<47:46:10, 168.27s/it, loss=0.0800, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1834/2855 [9:09:50<34:30:27, 121.67s/it, loss=0.0800, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1834/2855 [9:10:03<34:30:27, 121.67s/it, loss=0.1885, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1835/2855 [9:10:03<25:13:58, 89.06s/it, loss=0.1885, lr=2.71e-05] 
Epoch 2/8:  64%|██████▍   | 1835/2855 [9:20:25<25:13:58, 89.06s/it, loss=0.2649, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1836/2855 [9:20:25<70:29:19, 249.03s/it, loss=0.2649, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1836/2855 [9:20:37<70:29:19, 249.03s/it, loss=0.0648, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1837/2855 [9:20:37<50:19:48, 177.99s/it, loss=0.0648, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1837/2855 [9:20:50<50:19:48, 177.99s/it, loss=0.2212, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1838/2855 [9:20:50<36:18:22, 128.52s/it, loss=0.2212, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1838/2855 [9:21:03<36:18:22, 128.52s/it, loss=0.5006, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1839/2855 [9:21:03<26:29:18, 93.86s/it, loss=0.5006, lr=2.71e-05] 
Epoch 2/8:  64%|██████▍   | 1839/2855 [9:28:20<26:29:18, 93.86s/it, loss=0.0674, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1840/2855 [9:28:20<55:30:30, 196.88s/it, loss=0.0674, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1840/2855 [9:28:26<55:30:30, 196.88s/it, loss=0.0659, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1841/2855 [9:28:26<39:17:24, 139.49s/it, loss=0.0659, lr=2.71e-05]
Epoch 2/8:  64%|██████▍   | 1841/2855 [9:28:56<39:17:24, 139.49s/it, loss=0.0568, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1842/2855 [9:28:56<29:58:31, 106.53s/it, loss=0.0568, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1842/2855 [9:29:06<29:58:31, 106.53s/it, loss=0.2186, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1843/2855 [9:29:06<21:48:17, 77.57s/it, loss=0.2186, lr=2.71e-05] 
Epoch 2/8:  65%|██████▍   | 1843/2855 [9:33:22<21:48:17, 77.57s/it, loss=0.0552, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1844/2855 [9:33:22<36:50:35, 131.19s/it, loss=0.0552, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1844/2855 [9:33:29<36:50:35, 131.19s/it, loss=0.0348, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1845/2855 [9:33:29<26:21:43, 93.96s/it, loss=0.0348, lr=2.71e-05] 
Epoch 2/8:  65%|██████▍   | 1845/2855 [9:34:04<26:21:43, 93.96s/it, loss=0.1022, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1846/2855 [9:34:04<21:23:11, 76.30s/it, loss=0.1022, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1846/2855 [10:21:23<21:23:11, 76.30s/it, loss=0.1477, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1847/2855 [10:21:23<253:26:43, 905.16s/it, loss=0.1477, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1847/2855 [10:21:26<253:26:43, 905.16s/it, loss=0.0879, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1848/2855 [10:21:26<177:29:14, 634.51s/it, loss=0.0879, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1848/2855 [10:21:40<177:29:14, 634.51s/it, loss=0.2416, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1849/2855 [10:21:40<125:13:38, 448.13s/it, loss=0.2416, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1849/2855 [10:28:22<125:13:38, 448.13s/it, loss=0.0507, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1850/2855 [10:28:22<121:18:54, 434.56s/it, loss=0.0507, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1850/2855 [10:28:35<121:18:54, 434.56s/it, loss=0.1250, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1851/2855 [10:28:35<85:54:20, 308.03s/it, loss=0.1250, lr=2.71e-05] 
Epoch 2/8:  65%|██████▍   | 1851/2855 [10:33:23<85:54:20, 308.03s/it, loss=0.0559, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1852/2855 [10:33:23<84:07:42, 301.96s/it, loss=0.0559, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1852/2855 [10:33:26<84:07:42, 301.96s/it, loss=0.1785, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1853/2855 [10:33:26<59:05:52, 212.33s/it, loss=0.1785, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1853/2855 [10:33:39<59:05:52, 212.33s/it, loss=0.0733, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1854/2855 [10:33:39<42:26:18, 152.63s/it, loss=0.0733, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1854/2855 [10:33:53<42:26:18, 152.63s/it, loss=0.1823, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1855/2855 [10:33:53<30:48:53, 110.93s/it, loss=0.1823, lr=2.71e-05]
Epoch 2/8:  65%|██████▍   | 1855/2855 [10:34:06<30:48:53, 110.93s/it, loss=0.1860, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1856/2855 [10:34:06<22:36:47, 81.49s/it, loss=0.1860, lr=2.71e-05] 
Epoch 2/8:  65%|██████▌   | 1856/2855 [10:36:15<22:36:47, 81.49s/it, loss=0.1112, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1857/2855 [10:36:15<26:31:45, 95.70s/it, loss=0.1112, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1857/2855 [10:36:23<26:31:45, 95.70s/it, loss=0.1687, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1858/2855 [10:36:23<19:12:12, 69.34s/it, loss=0.1687, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1858/2855 [10:36:48<19:12:12, 69.34s/it, loss=0.1935, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1859/2855 [10:36:48<15:30:48, 56.07s/it, loss=0.1935, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1859/2855 [10:36:59<15:30:48, 56.07s/it, loss=0.1285, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1860/2855 [10:36:59<11:47:13, 42.65s/it, loss=0.1285, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1860/2855 [10:37:17<11:47:13, 42.65s/it, loss=0.0998, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1861/2855 [10:37:17<9:45:46, 35.36s/it, loss=0.0998, lr=2.71e-05] 
Epoch 2/8:  65%|██████▌   | 1861/2855 [10:37:24<9:45:46, 35.36s/it, loss=0.2119, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1862/2855 [10:37:24<7:20:40, 26.63s/it, loss=0.2119, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1862/2855 [10:37:37<7:20:40, 26.63s/it, loss=0.1755, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1863/2855 [10:37:37<6:15:11, 22.69s/it, loss=0.1755, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1863/2855 [10:37:50<6:15:11, 22.69s/it, loss=0.1461, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1864/2855 [10:37:50<5:25:29, 19.71s/it, loss=0.1461, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1864/2855 [10:38:02<5:25:29, 19.71s/it, loss=0.1137, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1865/2855 [10:38:02<4:46:38, 17.37s/it, loss=0.1137, lr=2.71e-05]
Epoch 2/8:  65%|██████▌   | 1865/2855 [11:22:26<4:46:38, 17.37s/it, loss=0.1496, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1866/2855 [11:22:26<222:52:31, 811.28s/it, loss=0.1496, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1866/2855 [11:22:32<222:52:31, 811.28s/it, loss=0.0012, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1867/2855 [11:22:32<156:22:00, 569.76s/it, loss=0.0012, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1867/2855 [11:22:54<156:22:00, 569.76s/it, loss=0.1234, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1868/2855 [11:22:54<111:11:12, 405.54s/it, loss=0.1234, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1868/2855 [11:23:07<111:11:12, 405.54s/it, loss=0.0390, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1869/2855 [11:23:07<78:50:29, 287.86s/it, loss=0.0390, lr=2.70e-05] 
Epoch 2/8:  65%|██████▌   | 1869/2855 [12:07:24<78:50:29, 287.86s/it, loss=0.0861, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1870/2855 [12:07:24<273:13:54, 998.61s/it, loss=0.0861, lr=2.70e-05]
Epoch 2/8:  65%|██████▌   | 1870/2855 [12:07:27<273:13:54, 998.61s/it, loss=0.1686, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1871/2855 [12:07:27<191:17:42, 699.86s/it, loss=0.1686, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1871/2855 [12:07:44<191:17:42, 699.86s/it, loss=0.2334, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1872/2855 [12:07:44<135:08:10, 494.90s/it, loss=0.2334, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1872/2855 [12:08:04<135:08:10, 494.90s/it, loss=0.0281, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1873/2855 [12:08:04<96:08:30, 352.45s/it, loss=0.0281, lr=2.70e-05] 
Epoch 2/8:  66%|██████▌   | 1873/2855 [12:15:26<96:08:30, 352.45s/it, loss=0.1042, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1874/2855 [12:15:26<103:24:24, 379.47s/it, loss=0.1042, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1874/2855 [12:15:42<103:24:24, 379.47s/it, loss=0.1768, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1875/2855 [12:15:42<73:32:57, 270.18s/it, loss=0.1768, lr=2.70e-05] 
Epoch 2/8:  66%|██████▌   | 1875/2855 [12:16:07<73:32:57, 270.18s/it, loss=0.1728, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1876/2855 [12:16:07<53:29:33, 196.70s/it, loss=0.1728, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1876/2855 [12:23:17<53:29:33, 196.70s/it, loss=0.2636, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1877/2855 [12:23:17<72:26:41, 266.67s/it, loss=0.2636, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1877/2855 [12:23:20<72:26:41, 266.67s/it, loss=0.1640, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1878/2855 [12:23:20<50:54:41, 187.60s/it, loss=0.1640, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1878/2855 [12:23:33<50:54:41, 187.60s/it, loss=0.1100, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1879/2855 [12:23:33<36:40:55, 135.30s/it, loss=0.1100, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1879/2855 [12:24:02<36:40:55, 135.30s/it, loss=0.3694, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1880/2855 [12:24:02<28:01:43, 103.49s/it, loss=0.3694, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1880/2855 [12:28:27<28:01:43, 103.49s/it, loss=0.1287, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1881/2855 [12:28:27<41:05:51, 151.90s/it, loss=0.1287, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1881/2855 [12:28:35<41:05:51, 151.90s/it, loss=0.1422, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1882/2855 [12:28:35<29:23:35, 108.75s/it, loss=0.1422, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1882/2855 [12:30:34<29:23:35, 108.75s/it, loss=0.2035, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1883/2855 [12:30:34<30:08:16, 111.62s/it, loss=0.2035, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1883/2855 [12:30:44<30:08:16, 111.62s/it, loss=0.0796, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1884/2855 [12:30:44<21:55:14, 81.27s/it, loss=0.0796, lr=2.70e-05] 
Epoch 2/8:  66%|██████▌   | 1884/2855 [12:31:09<21:55:14, 81.27s/it, loss=0.1091, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1885/2855 [12:31:09<17:19:36, 64.31s/it, loss=0.1091, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1885/2855 [12:31:17<17:19:36, 64.31s/it, loss=0.1808, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1886/2855 [12:31:17<12:47:34, 47.53s/it, loss=0.1808, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1886/2855 [12:51:47<12:47:34, 47.53s/it, loss=0.0882, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1887/2855 [12:51:47<108:09:43, 402.26s/it, loss=0.0882, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1887/2855 [12:51:53<108:09:43, 402.26s/it, loss=0.1339, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1888/2855 [12:51:53<76:04:14, 283.20s/it, loss=0.1339, lr=2.70e-05] 
Epoch 2/8:  66%|██████▌   | 1888/2855 [12:52:29<76:04:14, 283.20s/it, loss=0.1960, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1889/2855 [12:52:29<56:07:12, 209.14s/it, loss=0.1960, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1889/2855 [12:53:25<56:07:12, 209.14s/it, loss=0.2757, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1890/2855 [12:53:25<43:47:13, 163.35s/it, loss=0.2757, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1890/2855 [12:53:28<43:47:13, 163.35s/it, loss=0.2021, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1891/2855 [12:53:28<30:50:54, 115.20s/it, loss=0.2021, lr=2.70e-05]
Epoch 2/8:  66%|██████▌   | 1891/2855 [12:53:40<30:50:54, 115.20s/it, loss=0.1403, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1892/2855 [12:53:40<22:30:27, 84.14s/it, loss=0.1403, lr=2.70e-05] 
Epoch 2/8:  66%|██████▋   | 1892/2855 [13:24:17<22:30:27, 84.14s/it, loss=0.2201, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1893/2855 [13:24:17<163:01:57, 610.10s/it, loss=0.2201, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1893/2855 [13:24:21<163:01:57, 610.10s/it, loss=0.1957, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1894/2855 [13:24:21<114:15:56, 428.05s/it, loss=0.1957, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1894/2855 [13:24:34<114:15:56, 428.05s/it, loss=0.0958, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1895/2855 [13:24:34<80:56:39, 303.54s/it, loss=0.0958, lr=2.70e-05] 
Epoch 2/8:  66%|██████▋   | 1895/2855 [14:25:21<80:56:39, 303.54s/it, loss=0.2614, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1896/2855 [14:25:21<348:06:50, 1306.79s/it, loss=0.2614, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1896/2855 [14:25:25<348:06:50, 1306.79s/it, loss=0.1231, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1897/2855 [14:25:25<243:44:31, 915.94s/it, loss=0.1231, lr=2.70e-05] 
Epoch 2/8:  66%|██████▋   | 1897/2855 [14:25:40<243:44:31, 915.94s/it, loss=0.0788, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1898/2855 [14:25:40<171:37:46, 645.63s/it, loss=0.0788, lr=2.70e-05]
Epoch 2/8:  66%|██████▋   | 1898/2855 [14:25:53<171:37:46, 645.63s/it, loss=0.2523, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1899/2855 [14:25:53<121:03:13, 455.85s/it, loss=0.2523, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1899/2855 [14:26:04<121:03:13, 455.85s/it, loss=0.1047, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1900/2855 [14:26:04<85:29:48, 322.29s/it, loss=0.1047, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1900/2855 [14:27:13<85:29:48, 322.29s/it, loss=0.4537, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1901/2855 [14:27:13<65:17:53, 246.41s/it, loss=0.4537, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1901/2855 [14:27:22<65:17:53, 246.41s/it, loss=0.2512, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1902/2855 [14:27:22<46:21:31, 175.12s/it, loss=0.2512, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1902/2855 [14:28:28<46:21:31, 175.12s/it, loss=0.1958, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1903/2855 [14:28:28<37:39:11, 142.39s/it, loss=0.1958, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1903/2855 [14:28:41<37:39:11, 142.39s/it, loss=0.2247, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1904/2855 [14:28:41<27:22:19, 103.62s/it, loss=0.2247, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1904/2855 [14:28:54<27:22:19, 103.62s/it, loss=0.0467, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1905/2855 [14:28:54<20:09:52, 76.41s/it, loss=0.0467, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1905/2855 [14:29:07<20:09:52, 76.41s/it, loss=0.1127, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1906/2855 [14:29:07<15:08:27, 57.44s/it, loss=0.1127, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1906/2855 [14:29:56<15:08:27, 57.44s/it, loss=0.1332, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1907/2855 [14:29:56<14:24:22, 54.71s/it, loss=0.1332, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1907/2855 [14:29:59<14:24:22, 54.71s/it, loss=0.2712, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1908/2855 [14:29:59<10:18:10, 39.17s/it, loss=0.2712, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1908/2855 [14:30:11<10:18:10, 39.17s/it, loss=0.1383, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1909/2855 [14:30:11<8:12:58, 31.27s/it, loss=0.1383, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1909/2855 [14:30:24<8:12:58, 31.27s/it, loss=0.1257, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1910/2855 [14:30:24<6:46:13, 25.79s/it, loss=0.1257, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1910/2855 [15:12:20<6:46:13, 25.79s/it, loss=0.0041, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1911/2855 [15:12:20<202:38:16, 772.77s/it, loss=0.0041, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1911/2855 [15:12:23<202:38:16, 772.77s/it, loss=0.2199, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1912/2855 [15:12:23<141:55:19, 541.80s/it, loss=0.2199, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1912/2855 [15:12:50<141:55:19, 541.80s/it, loss=0.0626, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1913/2855 [15:12:50<101:23:27, 387.48s/it, loss=0.0626, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1913/2855 [15:13:01<101:23:27, 387.48s/it, loss=0.1711, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1914/2855 [15:13:01<71:44:35, 274.47s/it, loss=0.1711, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1914/2855 [15:17:16<71:44:35, 274.47s/it, loss=0.1196, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1915/2855 [15:17:16<70:08:42, 268.64s/it, loss=0.1196, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1915/2855 [15:17:30<70:08:42, 268.64s/it, loss=0.1705, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1916/2855 [15:17:30<50:05:34, 192.05s/it, loss=0.1705, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1916/2855 [15:26:19<50:05:34, 192.05s/it, loss=0.0925, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1917/2855 [15:26:19<76:23:08, 293.16s/it, loss=0.0925, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1917/2855 [15:26:21<76:23:08, 293.16s/it, loss=0.1559, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1918/2855 [15:26:21<53:36:37, 205.97s/it, loss=0.1559, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1918/2855 [15:26:27<53:36:37, 205.97s/it, loss=0.1005, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1919/2855 [15:26:27<37:56:34, 145.93s/it, loss=0.1005, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1919/2855 [15:26:41<37:56:34, 145.93s/it, loss=0.2641, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1920/2855 [15:26:41<27:35:11, 106.22s/it, loss=0.2641, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1920/2855 [15:26:53<27:35:11, 106.22s/it, loss=0.3181, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1921/2855 [15:26:53<20:14:12, 78.00s/it, loss=0.3181, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1921/2855 [15:27:05<20:14:12, 78.00s/it, loss=0.0991, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1922/2855 [15:27:05<15:05:27, 58.23s/it, loss=0.0991, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1922/2855 [16:04:18<15:05:27, 58.23s/it, loss=0.0521, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1923/2855 [16:04:18<184:00:27, 710.76s/it, loss=0.0521, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1923/2855 [16:04:25<184:00:27, 710.76s/it, loss=0.0949, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1924/2855 [16:04:25<129:12:09, 499.60s/it, loss=0.0949, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1924/2855 [16:04:38<129:12:09, 499.60s/it, loss=0.2765, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1925/2855 [16:04:38<91:21:35, 353.65s/it, loss=0.2765, lr=2.70e-05] 
Epoch 2/8:  67%|██████▋   | 1925/2855 [16:04:59<91:21:35, 353.65s/it, loss=0.0839, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1926/2855 [16:04:59<65:28:38, 253.73s/it, loss=0.0839, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1926/2855 [16:18:55<65:28:38, 253.73s/it, loss=0.2919, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1927/2855 [16:18:55<110:29:41, 428.64s/it, loss=0.2919, lr=2.70e-05]
Epoch 2/8:  67%|██████▋   | 1927/2855 [16:18:59<110:29:41, 428.64s/it, loss=0.2134, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1928/2855 [16:18:59<77:30:28, 301.00s/it, loss=0.2134, lr=2.70e-05] 
Epoch 2/8:  68%|██████▊   | 1928/2855 [16:22:24<77:30:28, 301.00s/it, loss=0.0997, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1929/2855 [16:22:24<70:02:15, 272.28s/it, loss=0.0997, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1929/2855 [16:22:27<70:02:15, 272.28s/it, loss=0.0850, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1930/2855 [16:22:27<49:11:29, 191.45s/it, loss=0.0850, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1930/2855 [16:22:29<49:11:29, 191.45s/it, loss=0.1846, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1931/2855 [16:22:29<34:35:30, 134.77s/it, loss=0.1846, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1931/2855 [16:22:32<34:35:30, 134.77s/it, loss=0.0675, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1932/2855 [16:22:32<24:23:10, 95.11s/it, loss=0.0675, lr=2.70e-05] 
Epoch 2/8:  68%|██████▊   | 1932/2855 [16:22:35<24:23:10, 95.11s/it, loss=0.0559, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1933/2855 [16:22:35<17:16:51, 67.47s/it, loss=0.0559, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1933/2855 [16:22:38<17:16:51, 67.47s/it, loss=0.1410, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1934/2855 [16:22:38<12:18:59, 48.14s/it, loss=0.1410, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1934/2855 [16:22:41<12:18:59, 48.14s/it, loss=0.1775, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1935/2855 [16:22:41<8:50:00, 34.57s/it, loss=0.1775, lr=2.70e-05] 
Epoch 2/8:  68%|██████▊   | 1935/2855 [16:22:44<8:50:00, 34.57s/it, loss=0.1247, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1936/2855 [16:22:44<6:24:16, 25.09s/it, loss=0.1247, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1936/2855 [16:22:47<6:24:16, 25.09s/it, loss=0.0320, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1937/2855 [16:22:47<4:42:53, 18.49s/it, loss=0.0320, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1937/2855 [16:22:50<4:42:53, 18.49s/it, loss=0.0717, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1938/2855 [16:22:50<3:32:12, 13.88s/it, loss=0.0717, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1938/2855 [16:22:53<3:32:12, 13.88s/it, loss=0.0614, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1939/2855 [16:22:53<2:42:04, 10.62s/it, loss=0.0614, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1939/2855 [16:22:56<2:42:04, 10.62s/it, loss=0.1017, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1940/2855 [16:22:56<2:07:22,  8.35s/it, loss=0.1017, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1940/2855 [16:22:59<2:07:22,  8.35s/it, loss=0.1330, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1941/2855 [16:22:59<1:43:13,  6.78s/it, loss=0.1330, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1941/2855 [16:23:02<1:43:13,  6.78s/it, loss=0.1599, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1942/2855 [16:23:02<1:26:01,  5.65s/it, loss=0.1599, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1942/2855 [16:23:05<1:26:01,  5.65s/it, loss=0.0722, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1943/2855 [16:23:05<1:14:19,  4.89s/it, loss=0.0722, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1943/2855 [16:23:09<1:14:19,  4.89s/it, loss=0.2592, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1944/2855 [16:23:09<1:08:36,  4.52s/it, loss=0.2592, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1944/2855 [16:23:13<1:08:36,  4.52s/it, loss=0.1980, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1945/2855 [16:23:13<1:05:11,  4.30s/it, loss=0.1980, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1945/2855 [16:23:17<1:05:11,  4.30s/it, loss=0.1649, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1946/2855 [16:23:17<1:05:26,  4.32s/it, loss=0.1649, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1946/2855 [16:23:20<1:05:26,  4.32s/it, loss=0.2331, lr=2.70e-05]
Epoch 2/8:  68%|██████▊   | 1947/2855 [16:23:20<59:49,  3.95s/it, loss=0.2331, lr=2.70e-05]  
Epoch 2/8:  68%|██████▊   | 1947/2855 [16:23:23<59:49,  3.95s/it, loss=0.1647, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1948/2855 [16:23:23<56:31,  3.74s/it, loss=0.1647, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1948/2855 [16:23:26<56:31,  3.74s/it, loss=0.1636, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1949/2855 [16:23:26<53:06,  3.52s/it, loss=0.1636, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1949/2855 [16:23:30<53:06,  3.52s/it, loss=0.1554, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1950/2855 [16:23:30<52:40,  3.49s/it, loss=0.1554, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1950/2855 [16:23:33<52:40,  3.49s/it, loss=0.0709, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1951/2855 [16:23:33<50:41,  3.36s/it, loss=0.0709, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1951/2855 [16:23:36<50:41,  3.36s/it, loss=0.2637, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1952/2855 [16:23:36<49:11,  3.27s/it, loss=0.2637, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1952/2855 [16:23:39<49:11,  3.27s/it, loss=0.0820, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1953/2855 [16:23:39<48:15,  3.21s/it, loss=0.0820, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1953/2855 [16:23:42<48:15,  3.21s/it, loss=0.1059, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1954/2855 [16:23:42<47:09,  3.14s/it, loss=0.1059, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1954/2855 [16:23:45<47:09,  3.14s/it, loss=0.1594, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1955/2855 [16:23:45<47:00,  3.13s/it, loss=0.1594, lr=2.69e-05]
Epoch 2/8:  68%|██████▊   | 1955/2855 [16:23:48<47:00,  3.13s/it, loss=0.1476, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1956/2855 [16:23:48<46:39,  3.11s/it, loss=0.1476, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1956/2855 [16:23:52<46:39,  3.11s/it, loss=0.1277, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1957/2855 [16:23:52<50:24,  3.37s/it, loss=0.1277, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1957/2855 [16:23:56<50:24,  3.37s/it, loss=0.1617, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1958/2855 [16:23:56<53:41,  3.59s/it, loss=0.1617, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1958/2855 [16:24:00<53:41,  3.59s/it, loss=0.1690, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1959/2855 [16:24:00<55:02,  3.69s/it, loss=0.1690, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1959/2855 [16:24:39<55:02,  3.69s/it, loss=0.0730, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1960/2855 [16:24:39<3:33:50, 14.34s/it, loss=0.0730, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1960/2855 [16:24:42<3:33:50, 14.34s/it, loss=0.2768, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1961/2855 [16:24:42<2:42:45, 10.92s/it, loss=0.2768, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1961/2855 [16:24:45<2:42:45, 10.92s/it, loss=0.0651, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1962/2855 [16:24:45<2:04:44,  8.38s/it, loss=0.0651, lr=2.69e-05]
Epoch 2/8:  69%|██████▊   | 1962/2855 [16:24:48<2:04:44,  8.38s/it, loss=0.1749, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1963/2855 [16:24:48<1:40:33,  6.76s/it, loss=0.1749, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1963/2855 [16:24:50<1:40:33,  6.76s/it, loss=0.1499, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1964/2855 [16:24:50<1:21:40,  5.50s/it, loss=0.1499, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1964/2855 [16:24:53<1:21:40,  5.50s/it, loss=0.2772, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1965/2855 [16:24:53<1:10:38,  4.76s/it, loss=0.2772, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1965/2855 [16:24:57<1:10:38,  4.76s/it, loss=0.1090, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1966/2855 [16:24:57<1:04:13,  4.33s/it, loss=0.1090, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1966/2855 [16:25:00<1:04:13,  4.33s/it, loss=0.1812, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1967/2855 [16:25:00<57:56,  3.91s/it, loss=0.1812, lr=2.69e-05]  
Epoch 2/8:  69%|██████▉   | 1967/2855 [16:25:03<57:56,  3.91s/it, loss=0.2055, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1968/2855 [16:25:03<55:10,  3.73s/it, loss=0.2055, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1968/2855 [16:25:06<55:10,  3.73s/it, loss=0.1071, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1969/2855 [16:25:06<51:45,  3.50s/it, loss=0.1071, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1969/2855 [16:25:09<51:45,  3.50s/it, loss=0.0941, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1970/2855 [16:25:09<50:35,  3.43s/it, loss=0.0941, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1970/2855 [16:25:13<50:35,  3.43s/it, loss=0.1847, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1971/2855 [16:25:13<50:10,  3.41s/it, loss=0.1847, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1971/2855 [16:25:15<50:10,  3.41s/it, loss=0.0836, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1972/2855 [16:25:15<48:17,  3.28s/it, loss=0.0836, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1972/2855 [16:25:19<48:17,  3.28s/it, loss=0.0964, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1973/2855 [16:25:19<47:05,  3.20s/it, loss=0.0964, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1973/2855 [16:25:22<47:05,  3.20s/it, loss=0.2401, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1974/2855 [16:25:22<47:14,  3.22s/it, loss=0.2401, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1974/2855 [16:25:25<47:14,  3.22s/it, loss=0.1331, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1975/2855 [16:25:25<46:57,  3.20s/it, loss=0.1331, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1975/2855 [16:25:28<46:57,  3.20s/it, loss=0.1581, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1976/2855 [16:25:28<46:42,  3.19s/it, loss=0.1581, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1976/2855 [16:25:31<46:42,  3.19s/it, loss=0.0987, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1977/2855 [16:25:31<46:11,  3.16s/it, loss=0.0987, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1977/2855 [16:25:34<46:11,  3.16s/it, loss=0.0952, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1978/2855 [16:25:34<45:46,  3.13s/it, loss=0.0952, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1978/2855 [16:25:37<45:46,  3.13s/it, loss=0.1401, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1979/2855 [16:25:37<46:01,  3.15s/it, loss=0.1401, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1979/2855 [16:25:41<46:01,  3.15s/it, loss=0.1191, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1980/2855 [16:25:41<46:07,  3.16s/it, loss=0.1191, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1980/2855 [16:25:44<46:07,  3.16s/it, loss=0.1761, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1981/2855 [16:25:44<45:50,  3.15s/it, loss=0.1761, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1981/2855 [16:25:47<45:50,  3.15s/it, loss=0.2265, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1982/2855 [16:25:47<45:52,  3.15s/it, loss=0.2265, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1982/2855 [16:25:50<45:52,  3.15s/it, loss=0.2933, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1983/2855 [16:25:50<46:01,  3.17s/it, loss=0.2933, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1983/2855 [16:25:54<46:01,  3.17s/it, loss=0.1101, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1984/2855 [16:25:54<49:09,  3.39s/it, loss=0.1101, lr=2.69e-05]
Epoch 2/8:  69%|██████▉   | 1984/2855 [16:26:05<49:09,  3.39s/it, loss=0.1902, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1985/2855 [16:26:05<1:20:51,  5.58s/it, loss=0.1902, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1985/2855 [16:26:08<1:20:51,  5.58s/it, loss=0.0621, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1986/2855 [16:26:08<1:09:34,  4.80s/it, loss=0.0621, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1986/2855 [16:26:11<1:09:34,  4.80s/it, loss=0.1730, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1987/2855 [16:26:11<1:02:41,  4.33s/it, loss=0.1730, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1987/2855 [16:26:14<1:02:41,  4.33s/it, loss=0.1957, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1988/2855 [16:26:14<57:16,  3.96s/it, loss=0.1957, lr=2.69e-05]  
Epoch 2/8:  70%|██████▉   | 1988/2855 [16:26:17<57:16,  3.96s/it, loss=0.1161, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1989/2855 [16:26:17<54:11,  3.75s/it, loss=0.1161, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1989/2855 [16:26:21<54:11,  3.75s/it, loss=0.1769, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1990/2855 [16:26:21<51:56,  3.60s/it, loss=0.1769, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1990/2855 [16:26:24<51:56,  3.60s/it, loss=0.1237, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1991/2855 [16:26:24<50:04,  3.48s/it, loss=0.1237, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1991/2855 [16:26:27<50:04,  3.48s/it, loss=0.1390, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1992/2855 [16:26:27<48:35,  3.38s/it, loss=0.1390, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1992/2855 [16:26:31<48:35,  3.38s/it, loss=0.1665, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1993/2855 [16:26:31<50:24,  3.51s/it, loss=0.1665, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1993/2855 [16:27:10<50:24,  3.51s/it, loss=0.0829, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1994/2855 [16:27:10<3:23:23, 14.17s/it, loss=0.0829, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1994/2855 [16:27:13<3:23:23, 14.17s/it, loss=0.0881, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1995/2855 [16:27:13<2:34:37, 10.79s/it, loss=0.0881, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1995/2855 [16:27:16<2:34:37, 10.79s/it, loss=0.2520, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1996/2855 [16:27:16<2:02:39,  8.57s/it, loss=0.2520, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1996/2855 [16:27:19<2:02:39,  8.57s/it, loss=0.1854, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1997/2855 [16:27:19<1:39:11,  6.94s/it, loss=0.1854, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1997/2855 [16:27:22<1:39:11,  6.94s/it, loss=0.1445, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1998/2855 [16:27:22<1:23:40,  5.86s/it, loss=0.1445, lr=2.69e-05]
Epoch 2/8:  70%|██████▉   | 1998/2855 [16:27:26<1:23:40,  5.86s/it, loss=0.0516, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 1999/2855 [16:27:26<1:13:17,  5.14s/it, loss=0.0516, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 1999/2855 [16:27:29<1:13:17,  5.14s/it, loss=0.1949, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2000/2855 [16:27:29<1:04:12,  4.51s/it, loss=0.1949, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2000/2855 [16:27:32<1:04:12,  4.51s/it, loss=0.1082, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2001/2855 [16:27:32<58:57,  4.14s/it, loss=0.1082, lr=2.69e-05]  
Epoch 2/8:  70%|███████   | 2001/2855 [16:27:36<58:57,  4.14s/it, loss=0.1501, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2002/2855 [16:27:36<55:43,  3.92s/it, loss=0.1501, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2002/2855 [16:27:39<55:43,  3.92s/it, loss=0.0306, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2003/2855 [16:27:39<52:18,  3.68s/it, loss=0.0306, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2003/2855 [16:27:42<52:18,  3.68s/it, loss=0.0708, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2004/2855 [16:27:42<50:43,  3.58s/it, loss=0.0708, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2004/2855 [16:27:44<50:43,  3.58s/it, loss=0.1130, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2005/2855 [16:27:44<44:51,  3.17s/it, loss=0.1130, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2005/2855 [16:27:48<44:51,  3.17s/it, loss=0.0728, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2006/2855 [16:27:48<45:13,  3.20s/it, loss=0.0728, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2006/2855 [16:27:50<45:13,  3.20s/it, loss=0.0236, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2007/2855 [16:27:50<42:22,  3.00s/it, loss=0.0236, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2007/2855 [16:27:55<42:22,  3.00s/it, loss=0.1795, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2008/2855 [16:27:55<51:36,  3.66s/it, loss=0.1795, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2008/2855 [16:27:59<51:36,  3.66s/it, loss=0.2001, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2009/2855 [16:27:59<50:18,  3.57s/it, loss=0.2001, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2009/2855 [16:28:02<50:18,  3.57s/it, loss=0.0912, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2010/2855 [16:28:02<49:44,  3.53s/it, loss=0.0912, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2010/2855 [16:28:05<49:44,  3.53s/it, loss=0.1144, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2011/2855 [16:28:05<48:25,  3.44s/it, loss=0.1144, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2011/2855 [16:28:09<48:25,  3.44s/it, loss=0.1883, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2012/2855 [16:28:09<48:10,  3.43s/it, loss=0.1883, lr=2.69e-05]
Epoch 2/8:  70%|███████   | 2012/2855 [16:28:12<48:10,  3.43s/it, loss=0.1410, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2013/2855 [16:28:12<48:17,  3.44s/it, loss=0.1410, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2013/2855 [16:28:16<48:17,  3.44s/it, loss=0.1626, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2014/2855 [16:28:16<48:33,  3.46s/it, loss=0.1626, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2014/2855 [16:28:19<48:33,  3.46s/it, loss=0.1926, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2015/2855 [16:28:19<48:01,  3.43s/it, loss=0.1926, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2015/2855 [16:28:23<48:01,  3.43s/it, loss=0.0367, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2016/2855 [16:28:23<49:17,  3.52s/it, loss=0.0367, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2016/2855 [16:28:26<49:17,  3.52s/it, loss=0.1308, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2017/2855 [16:28:26<48:34,  3.48s/it, loss=0.1308, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2017/2855 [16:28:29<48:34,  3.48s/it, loss=0.0857, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2018/2855 [16:28:29<47:24,  3.40s/it, loss=0.0857, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2018/2855 [16:28:33<47:24,  3.40s/it, loss=0.1799, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2019/2855 [16:28:33<47:24,  3.40s/it, loss=0.1799, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2019/2855 [16:28:36<47:24,  3.40s/it, loss=0.1990, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2020/2855 [16:28:36<46:41,  3.36s/it, loss=0.1990, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2020/2855 [16:28:39<46:41,  3.36s/it, loss=0.2449, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2021/2855 [16:28:39<46:39,  3.36s/it, loss=0.2449, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2021/2855 [16:28:43<46:39,  3.36s/it, loss=0.1789, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2022/2855 [16:28:43<46:51,  3.38s/it, loss=0.1789, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2022/2855 [16:28:47<46:51,  3.38s/it, loss=0.1074, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2023/2855 [16:28:47<47:57,  3.46s/it, loss=0.1074, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2023/2855 [16:28:50<47:57,  3.46s/it, loss=0.0341, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2024/2855 [16:28:50<47:15,  3.41s/it, loss=0.0341, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2024/2855 [16:28:53<47:15,  3.41s/it, loss=0.1144, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2025/2855 [16:28:53<47:36,  3.44s/it, loss=0.1144, lr=2.69e-05]
Epoch 2/8:  71%|█��█████   | 2025/2855 [16:28:57<47:36,  3.44s/it, loss=0.1924, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2026/2855 [16:28:57<47:12,  3.42s/it, loss=0.1924, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2026/2855 [16:29:00<47:12,  3.42s/it, loss=0.1859, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2027/2855 [16:29:00<46:36,  3.38s/it, loss=0.1859, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2027/2855 [16:29:03<46:36,  3.38s/it, loss=0.1942, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2028/2855 [16:29:03<46:50,  3.40s/it, loss=0.1942, lr=2.69e-05]
Epoch 2/8:  71%|███████   | 2028/2855 [16:29:07<46:50,  3.40s/it, loss=0.2834, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2029/2855 [16:29:07<46:53,  3.41s/it, loss=0.2834, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2029/2855 [16:29:10<46:53,  3.41s/it, loss=0.2574, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2030/2855 [16:29:10<46:04,  3.35s/it, loss=0.2574, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2030/2855 [16:29:13<46:04,  3.35s/it, loss=0.1414, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2031/2855 [16:29:13<46:12,  3.36s/it, loss=0.1414, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2031/2855 [16:29:17<46:12,  3.36s/it, loss=0.0897, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2032/2855 [16:29:17<46:13,  3.37s/it, loss=0.0897, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2032/2855 [16:29:20<46:13,  3.37s/it, loss=0.0330, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2033/2855 [16:29:20<45:36,  3.33s/it, loss=0.0330, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2033/2855 [16:29:23<45:36,  3.33s/it, loss=0.1008, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2034/2855 [16:29:23<45:48,  3.35s/it, loss=0.1008, lr=2.68e-05]
Epoch 2/8:  71%|███████   | 2034/2855 [16:29:27<45:48,  3.35s/it, loss=0.1748, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2035/2855 [16:29:27<45:38,  3.34s/it, loss=0.1748, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2035/2855 [16:29:30<45:38,  3.34s/it, loss=0.1817, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2036/2855 [16:29:30<44:36,  3.27s/it, loss=0.1817, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2036/2855 [16:29:33<44:36,  3.27s/it, loss=0.0794, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2037/2855 [16:29:33<44:58,  3.30s/it, loss=0.0794, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2037/2855 [16:29:37<44:58,  3.30s/it, loss=0.1165, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2038/2855 [16:29:37<44:46,  3.29s/it, loss=0.1165, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2038/2855 [16:29:40<44:46,  3.29s/it, loss=0.1404, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2039/2855 [16:29:40<44:04,  3.24s/it, loss=0.1404, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2039/2855 [16:29:43<44:04,  3.24s/it, loss=0.1277, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2040/2855 [16:29:43<44:02,  3.24s/it, loss=0.1277, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2040/2855 [16:29:46<44:02,  3.24s/it, loss=0.0919, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2041/2855 [16:29:46<43:58,  3.24s/it, loss=0.0919, lr=2.68e-05]
Epoch 2/8:  71%|███████▏  | 2041/2855 [16:29:50<43:58,  3.24s/it, loss=0.1647, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2042/2855 [16:29:50<46:02,  3.40s/it, loss=0.1647, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2042/2855 [16:29:53<46:02,  3.40s/it, loss=0.1559, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2043/2855 [16:29:53<45:45,  3.38s/it, loss=0.1559, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2043/2855 [16:29:57<45:45,  3.38s/it, loss=0.1837, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2044/2855 [16:29:57<45:25,  3.36s/it, loss=0.1837, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2044/2855 [16:30:00<45:25,  3.36s/it, loss=0.2253, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2045/2855 [16:30:00<45:25,  3.37s/it, loss=0.2253, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2045/2855 [16:30:03<45:25,  3.37s/it, loss=0.1428, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2046/2855 [16:30:03<45:14,  3.36s/it, loss=0.1428, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2046/2855 [16:30:10<45:14,  3.36s/it, loss=0.0703, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2047/2855 [16:30:10<58:31,  4.35s/it, loss=0.0703, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2047/2855 [16:30:13<58:31,  4.35s/it, loss=0.2222, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2048/2855 [16:30:13<54:19,  4.04s/it, loss=0.2222, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2048/2855 [16:30:17<54:19,  4.04s/it, loss=0.1012, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2049/2855 [16:30:17<51:03,  3.80s/it, loss=0.1012, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2049/2855 [16:30:20<51:03,  3.80s/it, loss=0.1573, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2050/2855 [16:30:20<49:06,  3.66s/it, loss=0.1573, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2050/2855 [16:30:23<49:06,  3.66s/it, loss=0.0896, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2051/2855 [16:30:23<47:38,  3.56s/it, loss=0.0896, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2051/2855 [16:30:26<47:38,  3.56s/it, loss=0.1536, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2052/2855 [16:30:26<46:29,  3.47s/it, loss=0.1536, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2052/2855 [16:30:30<46:29,  3.47s/it, loss=0.1171, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2053/2855 [16:30:30<45:14,  3.38s/it, loss=0.1171, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2053/2855 [16:30:33<45:14,  3.38s/it, loss=0.2045, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2054/2855 [16:30:33<44:59,  3.37s/it, loss=0.2045, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2054/2855 [16:30:36<44:59,  3.37s/it, loss=0.0617, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2055/2855 [16:30:36<44:30,  3.34s/it, loss=0.0617, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2055/2855 [16:30:40<44:30,  3.34s/it, loss=0.0675, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2056/2855 [16:30:40<44:47,  3.36s/it, loss=0.0675, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2056/2855 [16:30:43<44:47,  3.36s/it, loss=0.2062, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2057/2855 [16:30:43<44:57,  3.38s/it, loss=0.2062, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2057/2855 [16:30:46<44:57,  3.38s/it, loss=0.1696, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2058/2855 [16:30:46<43:44,  3.29s/it, loss=0.1696, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2058/2855 [16:30:50<43:44,  3.29s/it, loss=0.3430, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2059/2855 [16:30:50<43:53,  3.31s/it, loss=0.3430, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2059/2855 [16:30:53<43:53,  3.31s/it, loss=0.1557, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2060/2855 [16:30:53<44:22,  3.35s/it, loss=0.1557, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2060/2855 [16:30:56<44:22,  3.35s/it, loss=0.1679, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2061/2855 [16:30:56<44:39,  3.38s/it, loss=0.1679, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2061/2855 [16:31:00<44:39,  3.38s/it, loss=0.1178, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2062/2855 [16:31:00<44:49,  3.39s/it, loss=0.1178, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2062/2855 [16:31:03<44:49,  3.39s/it, loss=0.1703, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2063/2855 [16:31:03<45:15,  3.43s/it, loss=0.1703, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2063/2855 [16:31:07<45:15,  3.43s/it, loss=0.1784, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2064/2855 [16:31:07<44:34,  3.38s/it, loss=0.1784, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2064/2855 [16:31:10<44:34,  3.38s/it, loss=0.1330, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2065/2855 [16:31:10<44:26,  3.38s/it, loss=0.1330, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2065/2855 [16:31:13<44:26,  3.38s/it, loss=0.2448, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2066/2855 [16:31:13<44:17,  3.37s/it, loss=0.2448, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2066/2855 [16:31:16<44:17,  3.37s/it, loss=0.1982, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2067/2855 [16:31:16<43:06,  3.28s/it, loss=0.1982, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2067/2855 [16:31:20<43:06,  3.28s/it, loss=0.3035, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2068/2855 [16:31:20<43:20,  3.30s/it, loss=0.3035, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2068/2855 [16:31:23<43:20,  3.30s/it, loss=0.3335, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2069/2855 [16:31:23<44:04,  3.36s/it, loss=0.3335, lr=2.68e-05]
Epoch 2/8:  72%|███████▏  | 2069/2855 [16:31:27<44:04,  3.36s/it, loss=0.1387, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2070/2855 [16:31:27<44:25,  3.40s/it, loss=0.1387, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2070/2855 [16:31:30<44:25,  3.40s/it, loss=0.2033, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2071/2855 [16:31:30<43:50,  3.36s/it, loss=0.2033, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2071/2855 [16:31:33<43:50,  3.36s/it, loss=0.1802, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2072/2855 [16:31:33<44:12,  3.39s/it, loss=0.1802, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2072/2855 [16:31:37<44:12,  3.39s/it, loss=0.0759, lr=2.68e-05]
Epoch 2/8:  73%|███████��  | 2073/2855 [16:31:37<43:54,  3.37s/it, loss=0.0759, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2073/2855 [16:31:40<43:54,  3.37s/it, loss=0.1447, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2074/2855 [16:31:40<43:54,  3.37s/it, loss=0.1447, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2074/2855 [16:31:44<43:54,  3.37s/it, loss=0.0358, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2075/2855 [16:31:44<44:14,  3.40s/it, loss=0.0358, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2075/2855 [16:31:47<44:14,  3.40s/it, loss=0.2367, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2076/2855 [16:31:47<42:56,  3.31s/it, loss=0.2367, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2076/2855 [16:31:50<42:56,  3.31s/it, loss=0.2218, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2077/2855 [16:31:50<42:48,  3.30s/it, loss=0.2218, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2077/2855 [16:31:53<42:48,  3.30s/it, loss=0.2639, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2078/2855 [16:31:53<42:47,  3.30s/it, loss=0.2639, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2078/2855 [16:31:57<42:47,  3.30s/it, loss=0.0784, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2079/2855 [16:31:57<43:16,  3.35s/it, loss=0.0784, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2079/2855 [16:32:00<43:16,  3.35s/it, loss=0.2188, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2080/2855 [16:32:00<43:40,  3.38s/it, loss=0.2188, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2080/2855 [16:32:04<43:40,  3.38s/it, loss=0.2517, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2081/2855 [16:32:04<43:55,  3.40s/it, loss=0.2517, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2081/2855 [16:32:07<43:55,  3.40s/it, loss=0.2487, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2082/2855 [16:32:07<43:59,  3.41s/it, loss=0.2487, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2082/2855 [16:32:10<43:59,  3.41s/it, loss=0.0867, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2083/2855 [16:32:10<43:48,  3.40s/it, loss=0.0867, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2083/2855 [16:32:14<43:48,  3.40s/it, loss=0.2269, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2084/2855 [16:32:14<44:04,  3.43s/it, loss=0.2269, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2084/2855 [16:32:17<44:04,  3.43s/it, loss=0.0883, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2085/2855 [16:32:17<43:15,  3.37s/it, loss=0.0883, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2085/2855 [16:32:21<43:15,  3.37s/it, loss=0.1459, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2086/2855 [16:32:21<43:34,  3.40s/it, loss=0.1459, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2086/2855 [16:32:28<43:34,  3.40s/it, loss=0.0296, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2087/2855 [16:32:28<58:24,  4.56s/it, loss=0.0296, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2087/2855 [16:32:31<58:24,  4.56s/it, loss=0.1642, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2088/2855 [16:32:31<54:01,  4.23s/it, loss=0.1642, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2088/2855 [16:32:35<54:01,  4.23s/it, loss=0.1000, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2089/2855 [16:32:35<50:40,  3.97s/it, loss=0.1000, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2089/2855 [16:32:38<50:40,  3.97s/it, loss=0.2158, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2090/2855 [16:32:38<48:24,  3.80s/it, loss=0.2158, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2090/2855 [16:32:42<48:24,  3.80s/it, loss=0.1467, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2091/2855 [16:32:42<46:44,  3.67s/it, loss=0.1467, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2091/2855 [16:32:45<46:44,  3.67s/it, loss=0.2560, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2092/2855 [16:32:45<45:41,  3.59s/it, loss=0.2560, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2092/2855 [16:32:48<45:41,  3.59s/it, loss=0.0553, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2093/2855 [16:32:48<44:20,  3.49s/it, loss=0.0553, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2093/2855 [16:32:51<44:20,  3.49s/it, loss=0.1789, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2094/2855 [16:32:51<43:24,  3.42s/it, loss=0.1789, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2094/2855 [16:32:55<43:24,  3.42s/it, loss=0.0999, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2095/2855 [16:32:55<43:18,  3.42s/it, loss=0.0999, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2095/2855 [16:32:58<43:18,  3.42s/it, loss=0.2106, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2096/2855 [16:32:58<42:24,  3.35s/it, loss=0.2106, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2096/2855 [16:33:01<42:24,  3.35s/it, loss=0.1195, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2097/2855 [16:33:01<41:57,  3.32s/it, loss=0.1195, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2097/2855 [16:33:05<41:57,  3.32s/it, loss=0.2159, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2098/2855 [16:33:05<43:18,  3.43s/it, loss=0.2159, lr=2.68e-05]
Epoch 2/8:  73%|███████▎  | 2098/2855 [16:33:08<43:18,  3.43s/it, loss=0.0981, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2099/2855 [16:33:08<42:25,  3.37s/it, loss=0.0981, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2099/2855 [16:33:12<42:25,  3.37s/it, loss=0.1694, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2100/2855 [16:33:12<42:45,  3.40s/it, loss=0.1694, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2100/2855 [16:33:15<42:45,  3.40s/it, loss=0.0967, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2101/2855 [16:33:15<43:40,  3.48s/it, loss=0.0967, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2101/2855 [16:33:19<43:40,  3.48s/it, loss=0.0772, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2102/2855 [16:33:19<43:05,  3.43s/it, loss=0.0772, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2102/2855 [16:33:23<43:05,  3.43s/it, loss=0.1911, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2103/2855 [16:33:23<44:49,  3.58s/it, loss=0.1911, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2103/2855 [16:33:26<44:49,  3.58s/it, loss=0.1926, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2104/2855 [16:33:26<44:36,  3.56s/it, loss=0.1926, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2104/2855 [16:33:30<44:36,  3.56s/it, loss=0.0970, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2105/2855 [16:33:30<46:56,  3.75s/it, loss=0.0970, lr=2.68e-05]
Epoch 2/8:  74%|███████▎  | 2105/2855 [16:33:34<46:56,  3.75s/it, loss=0.0343, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2106/2855 [16:33:34<47:01,  3.77s/it, loss=0.0343, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2106/2855 [16:33:38<47:01,  3.77s/it, loss=0.0494, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2107/2855 [16:33:38<46:37,  3.74s/it, loss=0.0494, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2107/2855 [16:33:41<46:37,  3.74s/it, loss=0.1248, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2108/2855 [16:33:41<44:55,  3.61s/it, loss=0.1248, lr=2.68e-05]
Epoch 2/8:  74%|███████▍  | 2108/2855 [16:33:44<44:55,  3.61s/it, loss=0.1428, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2109/2855 [16:33:44<43:48,  3.52s/it, loss=0.1428, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2109/2855 [16:33:48<43:48,  3.52s/it, loss=0.2153, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2110/2855 [16:33:48<43:41,  3.52s/it, loss=0.2153, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2110/2855 [16:33:51<43:41,  3.52s/it, loss=0.1793, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2111/2855 [16:33:51<43:29,  3.51s/it, loss=0.1793, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2111/2855 [16:33:55<43:29,  3.51s/it, loss=0.1439, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2112/2855 [16:33:55<43:04,  3.48s/it, loss=0.1439, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2112/2855 [16:33:58<43:04,  3.48s/it, loss=0.1763, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2113/2855 [16:33:58<42:06,  3.40s/it, loss=0.1763, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2113/2855 [16:34:01<42:06,  3.40s/it, loss=0.1314, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2114/2855 [16:34:01<42:00,  3.40s/it, loss=0.1314, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2114/2855 [16:34:05<42:00,  3.40s/it, loss=0.1542, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2115/2855 [16:34:05<42:01,  3.41s/it, loss=0.1542, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2115/2855 [16:34:08<42:01,  3.41s/it, loss=0.2382, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2116/2855 [16:34:08<42:13,  3.43s/it, loss=0.2382, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2116/2855 [16:34:12<42:13,  3.43s/it, loss=0.1016, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2117/2855 [16:34:12<42:09,  3.43s/it, loss=0.1016, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2117/2855 [16:34:15<42:09,  3.43s/it, loss=0.1920, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2118/2855 [16:34:15<42:19,  3.45s/it, loss=0.1920, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2118/2855 [16:34:19<42:19,  3.45s/it, loss=0.1206, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2119/2855 [16:34:19<41:43,  3.40s/it, loss=0.1206, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2119/2855 [16:34:22<41:43,  3.40s/it, loss=0.3111, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2120/2855 [16:34:22<41:44,  3.41s/it, loss=0.3111, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2120/2855 [16:34:25<41:44,  3.41s/it, loss=0.0769, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2121/2855 [16:34:25<41:42,  3.41s/it, loss=0.0769, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2121/2855 [16:34:29<41:42,  3.41s/it, loss=0.2549, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2122/2855 [16:34:29<41:08,  3.37s/it, loss=0.2549, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2122/2855 [16:34:32<41:08,  3.37s/it, loss=0.0873, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2123/2855 [16:34:32<40:28,  3.32s/it, loss=0.0873, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2123/2855 [16:34:35<40:28,  3.32s/it, loss=0.2543, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2124/2855 [16:34:35<41:06,  3.37s/it, loss=0.2543, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2124/2855 [16:34:39<41:06,  3.37s/it, loss=0.0278, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2125/2855 [16:34:39<40:20,  3.32s/it, loss=0.0278, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2125/2855 [16:34:42<40:20,  3.32s/it, loss=0.1691, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2126/2855 [16:34:42<41:00,  3.38s/it, loss=0.1691, lr=2.67e-05]
Epoch 2/8:  74%|███████▍  | 2126/2855 [16:34:51<41:00,  3.38s/it, loss=0.2012, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2127/2855 [16:34:51<1:00:13,  4.96s/it, loss=0.2012, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2127/2855 [16:34:54<1:00:13,  4.96s/it, loss=0.1761, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2128/2855 [16:34:54<54:21,  4.49s/it, loss=0.1761, lr=2.67e-05]  
Epoch 2/8:  75%|███████▍  | 2128/2855 [16:34:57<54:21,  4.49s/it, loss=0.1543, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2129/2855 [16:34:57<49:47,  4.11s/it, loss=0.1543, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2129/2855 [16:35:01<49:47,  4.11s/it, loss=0.1610, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2130/2855 [16:35:01<47:03,  3.89s/it, loss=0.1610, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2130/2855 [16:35:04<47:03,  3.89s/it, loss=0.1437, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2131/2855 [16:35:04<44:59,  3.73s/it, loss=0.1437, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2131/2855 [16:35:08<44:59,  3.73s/it, loss=0.0336, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2132/2855 [16:35:08<43:46,  3.63s/it, loss=0.0336, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2132/2855 [16:35:11<43:46,  3.63s/it, loss=0.3799, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2133/2855 [16:35:11<44:33,  3.70s/it, loss=0.3799, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2133/2855 [16:35:15<44:33,  3.70s/it, loss=0.1427, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2134/2855 [16:35:15<42:33,  3.54s/it, loss=0.1427, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2134/2855 [16:35:18<42:33,  3.54s/it, loss=0.2405, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2135/2855 [16:35:18<41:56,  3.50s/it, loss=0.2405, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2135/2855 [16:35:21<41:56,  3.50s/it, loss=0.0907, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2136/2855 [16:35:21<42:08,  3.52s/it, loss=0.0907, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2136/2855 [16:35:25<42:08,  3.52s/it, loss=0.0687, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2137/2855 [16:35:25<41:40,  3.48s/it, loss=0.0687, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2137/2855 [16:35:28<41:40,  3.48s/it, loss=0.1447, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2138/2855 [16:35:28<41:05,  3.44s/it, loss=0.1447, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2138/2855 [16:35:32<41:05,  3.44s/it, loss=0.0457, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2139/2855 [16:35:32<41:24,  3.47s/it, loss=0.0457, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2139/2855 [16:35:35<41:24,  3.47s/it, loss=0.1814, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2140/2855 [16:35:35<41:32,  3.49s/it, loss=0.1814, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2140/2855 [16:35:39<41:32,  3.49s/it, loss=0.1347, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2141/2855 [16:35:39<41:55,  3.52s/it, loss=0.1347, lr=2.67e-05]
Epoch 2/8:  75%|███████▍  | 2141/2855 [16:35:43<41:55,  3.52s/it, loss=0.1341, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2142/2855 [16:35:43<42:25,  3.57s/it, loss=0.1341, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2142/2855 [16:35:46<42:25,  3.57s/it, loss=0.3418, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2143/2855 [16:35:46<42:26,  3.58s/it, loss=0.3418, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2143/2855 [16:35:50<42:26,  3.58s/it, loss=0.0979, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2144/2855 [16:35:50<41:57,  3.54s/it, loss=0.0979, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2144/2855 [16:35:54<41:57,  3.54s/it, loss=0.0651, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2145/2855 [16:35:54<43:57,  3.71s/it, loss=0.0651, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2145/2855 [16:35:58<43:57,  3.71s/it, loss=0.1010, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2146/2855 [16:35:58<46:56,  3.97s/it, loss=0.1010, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2146/2855 [16:36:03<46:56,  3.97s/it, loss=0.2796, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2147/2855 [16:36:03<47:49,  4.05s/it, loss=0.2796, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2147/2855 [16:36:06<47:49,  4.05s/it, loss=0.1950, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2148/2855 [16:36:06<46:49,  3.97s/it, loss=0.1950, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2148/2855 [16:36:10<46:49,  3.97s/it, loss=0.2875, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2149/2855 [16:36:10<45:10,  3.84s/it, loss=0.2875, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2149/2855 [16:36:13<45:10,  3.84s/it, loss=0.0656, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2150/2855 [16:36:13<43:33,  3.71s/it, loss=0.0656, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2150/2855 [16:36:17<43:33,  3.71s/it, loss=0.1273, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2151/2855 [16:36:17<43:11,  3.68s/it, loss=0.1273, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2151/2855 [16:36:20<43:11,  3.68s/it, loss=0.1385, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2152/2855 [16:36:20<41:54,  3.58s/it, loss=0.1385, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2152/2855 [16:36:24<41:54,  3.58s/it, loss=0.2182, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2153/2855 [16:36:24<41:07,  3.51s/it, loss=0.2182, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2153/2855 [16:36:27<41:07,  3.51s/it, loss=0.1406, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2154/2855 [16:36:27<40:39,  3.48s/it, loss=0.1406, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2154/2855 [16:36:30<40:39,  3.48s/it, loss=0.0910, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2155/2855 [16:36:30<39:54,  3.42s/it, loss=0.0910, lr=2.67e-05]
Epoch 2/8:  75%|███████▌  | 2155/2855 [16:36:34<39:54,  3.42s/it, loss=0.2403, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2156/2855 [16:36:34<39:07,  3.36s/it, loss=0.2403, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2156/2855 [16:36:37<39:07,  3.36s/it, loss=0.0854, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2157/2855 [16:36:37<39:25,  3.39s/it, loss=0.0854, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2157/2855 [16:36:40<39:25,  3.39s/it, loss=0.0919, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2158/2855 [16:36:40<38:50,  3.34s/it, loss=0.0919, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2158/2855 [16:36:44<38:50,  3.34s/it, loss=0.1379, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2159/2855 [16:36:44<39:01,  3.36s/it, loss=0.1379, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2159/2855 [16:36:47<39:01,  3.36s/it, loss=0.1587, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2160/2855 [16:36:47<39:15,  3.39s/it, loss=0.1587, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2160/2855 [16:36:50<39:15,  3.39s/it, loss=0.1515, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2161/2855 [16:36:50<39:03,  3.38s/it, loss=0.1515, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2161/2855 [16:36:54<39:03,  3.38s/it, loss=0.1788, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2162/2855 [16:36:54<38:10,  3.31s/it, loss=0.1788, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2162/2855 [16:36:57<38:10,  3.31s/it, loss=0.0857, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2163/2855 [16:36:57<38:19,  3.32s/it, loss=0.0857, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2163/2855 [16:37:00<38:19,  3.32s/it, loss=0.2035, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2164/2855 [16:37:00<38:03,  3.30s/it, loss=0.2035, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2164/2855 [16:37:03<38:03,  3.30s/it, loss=0.0744, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2165/2855 [16:37:03<37:45,  3.28s/it, loss=0.0744, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2165/2855 [16:37:10<37:45,  3.28s/it, loss=0.0937, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2166/2855 [16:37:10<47:44,  4.16s/it, loss=0.0937, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2166/2855 [16:37:13<47:44,  4.16s/it, loss=0.0815, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2167/2855 [16:37:13<45:29,  3.97s/it, loss=0.0815, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2167/2855 [16:37:17<45:29,  3.97s/it, loss=0.1368, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2168/2855 [16:37:17<45:11,  3.95s/it, loss=0.1368, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2168/2855 [16:37:21<45:11,  3.95s/it, loss=0.1118, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2169/2855 [16:37:21<43:45,  3.83s/it, loss=0.1118, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2169/2855 [16:37:24<43:45,  3.83s/it, loss=0.1787, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2170/2855 [16:37:24<43:04,  3.77s/it, loss=0.1787, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2170/2855 [16:37:28<43:04,  3.77s/it, loss=0.1081, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2171/2855 [16:37:28<43:17,  3.80s/it, loss=0.1081, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2171/2855 [16:37:32<43:17,  3.80s/it, loss=0.1302, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2172/2855 [16:37:32<42:45,  3.76s/it, loss=0.1302, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2172/2855 [16:37:35<42:45,  3.76s/it, loss=0.1447, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2173/2855 [16:37:35<42:38,  3.75s/it, loss=0.1447, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2173/2855 [16:37:39<42:38,  3.75s/it, loss=0.0760, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2174/2855 [16:37:39<41:25,  3.65s/it, loss=0.0760, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2174/2855 [16:37:43<41:25,  3.65s/it, loss=0.2173, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2175/2855 [16:37:43<41:24,  3.65s/it, loss=0.2173, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2175/2855 [16:37:46<41:24,  3.65s/it, loss=0.0036, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2176/2855 [16:37:46<40:31,  3.58s/it, loss=0.0036, lr=2.67e-05]
Epoch 2/8:  76%|███████▌  | 2176/2855 [16:37:49<40:31,  3.58s/it, loss=0.0391, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2177/2855 [16:37:49<39:44,  3.52s/it, loss=0.0391, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2177/2855 [16:37:53<39:44,  3.52s/it, loss=0.2035, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2178/2855 [16:37:53<40:02,  3.55s/it, loss=0.2035, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2178/2855 [16:37:57<40:02,  3.55s/it, loss=0.0819, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2179/2855 [16:37:57<40:03,  3.56s/it, loss=0.0819, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2179/2855 [16:38:00<40:03,  3.56s/it, loss=0.1102, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2180/2855 [16:38:00<39:13,  3.49s/it, loss=0.1102, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2180/2855 [16:38:03<39:13,  3.49s/it, loss=0.0975, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2181/2855 [16:38:03<39:39,  3.53s/it, loss=0.0975, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2181/2855 [16:38:07<39:39,  3.53s/it, loss=0.1857, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2182/2855 [16:38:07<39:41,  3.54s/it, loss=0.1857, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2182/2855 [16:38:10<39:41,  3.54s/it, loss=0.1963, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2183/2855 [16:38:10<39:07,  3.49s/it, loss=0.1963, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2183/2855 [16:38:14<39:07,  3.49s/it, loss=0.1340, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2184/2855 [16:38:14<38:17,  3.42s/it, loss=0.1340, lr=2.67e-05]
Epoch 2/8:  76%|███████▋  | 2184/2855 [16:38:17<38:17,  3.42s/it, loss=0.1144, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2185/2855 [16:38:17<38:26,  3.44s/it, loss=0.1144, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2185/2855 [16:38:20<38:26,  3.44s/it, loss=0.1592, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2186/2855 [16:38:20<37:44,  3.39s/it, loss=0.1592, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2186/2855 [16:38:24<37:44,  3.39s/it, loss=0.0858, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2187/2855 [16:38:24<37:48,  3.40s/it, loss=0.0858, lr=2.67e-05]
Epoch 2/8:  77%|███████▋  | 2187/2855 [16:38:28<37:48,  3.40s/it, loss=0.1129, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2188/2855 [16:38:28<39:10,  3.52s/it, loss=0.1129, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2188/2855 [16:38:31<39:10,  3.52s/it, loss=0.1698, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2189/2855 [16:38:31<38:17,  3.45s/it, loss=0.1698, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2189/2855 [16:38:34<38:17,  3.45s/it, loss=0.1639, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2190/2855 [16:38:34<37:35,  3.39s/it, loss=0.1639, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2190/2855 [16:38:38<37:35,  3.39s/it, loss=0.0803, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2191/2855 [16:38:38<38:38,  3.49s/it, loss=0.0803, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2191/2855 [16:38:41<38:38,  3.49s/it, loss=0.1356, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2192/2855 [16:38:41<38:10,  3.45s/it, loss=0.1356, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2192/2855 [16:38:45<38:10,  3.45s/it, loss=0.1457, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2193/2855 [16:38:45<37:29,  3.40s/it, loss=0.1457, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2193/2855 [16:38:48<37:29,  3.40s/it, loss=0.1168, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2194/2855 [16:38:48<37:48,  3.43s/it, loss=0.1168, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2194/2855 [16:38:52<37:48,  3.43s/it, loss=0.1330, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2195/2855 [16:38:52<38:02,  3.46s/it, loss=0.1330, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2195/2855 [16:38:55<38:02,  3.46s/it, loss=0.0609, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2196/2855 [16:38:55<37:27,  3.41s/it, loss=0.0609, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2196/2855 [16:38:58<37:27,  3.41s/it, loss=0.1118, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2197/2855 [16:38:58<37:56,  3.46s/it, loss=0.1118, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2197/2855 [16:39:02<37:56,  3.46s/it, loss=0.0954, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2198/2855 [16:39:02<37:39,  3.44s/it, loss=0.0954, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2198/2855 [16:39:05<37:39,  3.44s/it, loss=0.1049, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2199/2855 [16:39:05<38:11,  3.49s/it, loss=0.1049, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2199/2855 [16:39:09<38:11,  3.49s/it, loss=0.2004, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2200/2855 [16:39:09<38:20,  3.51s/it, loss=0.2004, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2200/2855 [16:39:12<38:20,  3.51s/it, loss=0.2500, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2201/2855 [16:39:12<38:00,  3.49s/it, loss=0.2500, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2201/2855 [16:39:16<38:00,  3.49s/it, loss=0.1281, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2202/2855 [16:39:16<37:18,  3.43s/it, loss=0.1281, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2202/2855 [16:39:19<37:18,  3.43s/it, loss=0.0644, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2203/2855 [16:39:19<37:12,  3.42s/it, loss=0.0644, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2203/2855 [16:39:23<37:12,  3.42s/it, loss=0.1821, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2204/2855 [16:39:23<37:28,  3.45s/it, loss=0.1821, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2204/2855 [16:39:26<37:28,  3.45s/it, loss=0.2356, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2205/2855 [16:39:26<36:55,  3.41s/it, loss=0.2356, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2205/2855 [16:39:34<36:55,  3.41s/it, loss=0.1743, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2206/2855 [16:39:34<51:29,  4.76s/it, loss=0.1743, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2206/2855 [16:39:37<51:29,  4.76s/it, loss=0.2173, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2207/2855 [16:39:37<47:08,  4.36s/it, loss=0.2173, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2207/2855 [16:39:40<47:08,  4.36s/it, loss=0.2673, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2208/2855 [16:39:40<42:59,  3.99s/it, loss=0.2673, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2208/2855 [16:39:44<42:59,  3.99s/it, loss=0.2643, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2209/2855 [16:39:44<40:51,  3.80s/it, loss=0.2643, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2209/2855 [16:39:47<40:51,  3.80s/it, loss=0.1343, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2210/2855 [16:39:47<38:00,  3.54s/it, loss=0.1343, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2210/2855 [16:39:50<38:00,  3.54s/it, loss=0.2049, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2211/2855 [16:39:50<35:31,  3.31s/it, loss=0.2049, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2211/2855 [16:39:53<35:31,  3.31s/it, loss=0.0964, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2212/2855 [16:39:53<36:26,  3.40s/it, loss=0.0964, lr=2.66e-05]
Epoch 2/8:  77%|███████▋  | 2212/2855 [16:39:56<36:26,  3.40s/it, loss=0.1214, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2213/2855 [16:39:56<35:36,  3.33s/it, loss=0.1214, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2213/2855 [16:40:00<35:36,  3.33s/it, loss=0.1017, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2214/2855 [16:40:00<35:40,  3.34s/it, loss=0.1017, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2214/2855 [16:40:03<35:40,  3.34s/it, loss=0.2164, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2215/2855 [16:40:03<35:58,  3.37s/it, loss=0.2164, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2215/2855 [16:40:06<35:58,  3.37s/it, loss=0.1881, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2216/2855 [16:40:06<35:37,  3.34s/it, loss=0.1881, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2216/2855 [16:40:10<35:37,  3.34s/it, loss=0.0651, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2217/2855 [16:40:10<35:27,  3.33s/it, loss=0.0651, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2217/2855 [16:40:13<35:27,  3.33s/it, loss=0.1567, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2218/2855 [16:40:13<35:38,  3.36s/it, loss=0.1567, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2218/2855 [16:40:16<35:38,  3.36s/it, loss=0.1872, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2219/2855 [16:40:16<35:33,  3.35s/it, loss=0.1872, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2219/2855 [16:40:20<35:33,  3.35s/it, loss=0.1984, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2220/2855 [16:40:20<34:49,  3.29s/it, loss=0.1984, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2220/2855 [16:40:23<34:49,  3.29s/it, loss=0.0055, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2221/2855 [16:40:23<35:59,  3.41s/it, loss=0.0055, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2221/2855 [16:40:27<35:59,  3.41s/it, loss=0.1505, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2222/2855 [16:40:27<35:36,  3.38s/it, loss=0.1505, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2222/2855 [16:40:30<35:36,  3.38s/it, loss=0.1576, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2223/2855 [16:40:30<34:41,  3.29s/it, loss=0.1576, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2223/2855 [16:40:33<34:41,  3.29s/it, loss=0.2142, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2224/2855 [16:40:33<34:46,  3.31s/it, loss=0.2142, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2224/2855 [16:40:36<34:46,  3.31s/it, loss=0.0690, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2225/2855 [16:40:36<34:31,  3.29s/it, loss=0.0690, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2225/2855 [16:40:39<34:31,  3.29s/it, loss=0.0945, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2226/2855 [16:40:39<32:31,  3.10s/it, loss=0.0945, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2226/2855 [16:40:42<32:31,  3.10s/it, loss=0.1381, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2227/2855 [16:40:42<32:20,  3.09s/it, loss=0.1381, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2227/2855 [16:40:45<32:20,  3.09s/it, loss=0.1511, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2228/2855 [16:40:45<30:42,  2.94s/it, loss=0.1511, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2228/2855 [16:40:47<30:42,  2.94s/it, loss=0.2985, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2229/2855 [16:40:47<29:37,  2.84s/it, loss=0.2985, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2229/2855 [16:40:50<29:37,  2.84s/it, loss=0.1746, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2230/2855 [16:40:50<29:33,  2.84s/it, loss=0.1746, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2230/2855 [16:40:54<29:33,  2.84s/it, loss=0.0842, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2231/2855 [16:40:54<32:17,  3.10s/it, loss=0.0842, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2231/2855 [16:40:57<32:17,  3.10s/it, loss=0.1492, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2232/2855 [16:40:57<32:21,  3.12s/it, loss=0.1492, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2232/2855 [16:41:00<32:21,  3.12s/it, loss=0.1425, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2233/2855 [16:41:00<33:46,  3.26s/it, loss=0.1425, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2233/2855 [16:41:04<33:46,  3.26s/it, loss=0.3047, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2234/2855 [16:41:04<34:24,  3.32s/it, loss=0.3047, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2234/2855 [16:41:07<34:24,  3.32s/it, loss=0.0294, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2235/2855 [16:41:07<34:34,  3.35s/it, loss=0.0294, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2235/2855 [16:41:10<34:34,  3.35s/it, loss=0.2253, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2236/2855 [16:41:10<33:54,  3.29s/it, loss=0.2253, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2236/2855 [16:41:14<33:54,  3.29s/it, loss=0.1530, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2237/2855 [16:41:14<35:54,  3.49s/it, loss=0.1530, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2237/2855 [16:41:18<35:54,  3.49s/it, loss=0.2631, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2238/2855 [16:41:18<35:38,  3.47s/it, loss=0.2631, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2238/2855 [16:41:21<35:38,  3.47s/it, loss=0.1160, lr=2.66e-05]
Epoch 2/8:  78%|���██████▊  | 2239/2855 [16:41:21<35:08,  3.42s/it, loss=0.1160, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2239/2855 [16:41:25<35:08,  3.42s/it, loss=0.1652, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2240/2855 [16:41:25<35:16,  3.44s/it, loss=0.1652, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2240/2855 [16:41:28<35:16,  3.44s/it, loss=0.1263, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2241/2855 [16:41:28<35:14,  3.44s/it, loss=0.1263, lr=2.66e-05]
Epoch 2/8:  78%|███████▊  | 2241/2855 [16:41:31<35:14,  3.44s/it, loss=0.0930, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2242/2855 [16:41:31<34:30,  3.38s/it, loss=0.0930, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2242/2855 [16:41:34<34:30,  3.38s/it, loss=0.2566, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2243/2855 [16:41:34<33:42,  3.30s/it, loss=0.2566, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2243/2855 [16:41:38<33:42,  3.30s/it, loss=0.1396, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2244/2855 [16:41:38<33:45,  3.31s/it, loss=0.1396, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2244/2855 [16:41:41<33:45,  3.31s/it, loss=0.0707, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2245/2855 [16:41:41<33:55,  3.34s/it, loss=0.0707, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2245/2855 [16:41:45<33:55,  3.34s/it, loss=0.1285, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2246/2855 [16:41:45<34:33,  3.40s/it, loss=0.1285, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2246/2855 [16:41:48<34:33,  3.40s/it, loss=0.1641, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2247/2855 [16:41:48<34:31,  3.41s/it, loss=0.1641, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2247/2855 [16:41:52<34:31,  3.41s/it, loss=0.3410, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2248/2855 [16:41:52<34:24,  3.40s/it, loss=0.3410, lr=2.66e-05]
Epoch 2/8:  79%|███████▊  | 2248/2855 [16:41:55<34:24,  3.40s/it, loss=0.0791, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2249/2855 [16:41:55<34:33,  3.42s/it, loss=0.0791, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2249/2855 [16:41:59<34:33,  3.42s/it, loss=0.1294, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2250/2855 [16:41:59<35:04,  3.48s/it, loss=0.1294, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2250/2855 [16:42:02<35:04,  3.48s/it, loss=0.0972, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2251/2855 [16:42:02<34:42,  3.45s/it, loss=0.0972, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2251/2855 [16:42:05<34:42,  3.45s/it, loss=0.0839, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2252/2855 [16:42:05<34:09,  3.40s/it, loss=0.0839, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2252/2855 [16:42:09<34:09,  3.40s/it, loss=0.1262, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2253/2855 [16:42:09<34:14,  3.41s/it, loss=0.1262, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2253/2855 [16:42:12<34:14,  3.41s/it, loss=0.2014, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2254/2855 [16:42:12<33:59,  3.39s/it, loss=0.2014, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2254/2855 [16:42:16<33:59,  3.39s/it, loss=0.0830, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2255/2855 [16:42:16<34:09,  3.42s/it, loss=0.0830, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2255/2855 [16:42:19<34:09,  3.42s/it, loss=0.1240, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2256/2855 [16:42:19<34:03,  3.41s/it, loss=0.1240, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2256/2855 [16:42:22<34:03,  3.41s/it, loss=0.0371, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2257/2855 [16:42:22<33:33,  3.37s/it, loss=0.0371, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2257/2855 [16:42:25<33:33,  3.37s/it, loss=0.4067, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2258/2855 [16:42:25<33:04,  3.32s/it, loss=0.4067, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2258/2855 [16:42:29<33:04,  3.32s/it, loss=0.1875, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2259/2855 [16:42:29<33:27,  3.37s/it, loss=0.1875, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2259/2855 [16:42:32<33:27,  3.37s/it, loss=0.1231, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2260/2855 [16:42:32<32:40,  3.30s/it, loss=0.1231, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2260/2855 [16:42:35<32:40,  3.30s/it, loss=0.2167, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2261/2855 [16:42:35<32:49,  3.32s/it, loss=0.2167, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2261/2855 [16:42:39<32:49,  3.32s/it, loss=0.1640, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2262/2855 [16:42:39<33:24,  3.38s/it, loss=0.1640, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2262/2855 [16:42:42<33:24,  3.38s/it, loss=0.0203, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2263/2855 [16:42:42<32:52,  3.33s/it, loss=0.0203, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2263/2855 [16:42:46<32:52,  3.33s/it, loss=0.2009, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2264/2855 [16:42:46<33:04,  3.36s/it, loss=0.2009, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2264/2855 [16:42:49<33:04,  3.36s/it, loss=0.2131, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2265/2855 [16:42:49<33:26,  3.40s/it, loss=0.2131, lr=2.66e-05]
Epoch 2/8:  79%|███████▉  | 2265/2855 [16:42:52<33:26,  3.40s/it, loss=0.2327, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2266/2855 [16:42:52<33:21,  3.40s/it, loss=0.2327, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2266/2855 [16:42:56<33:21,  3.40s/it, loss=0.0988, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2267/2855 [16:42:56<32:42,  3.34s/it, loss=0.0988, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2267/2855 [16:42:59<32:42,  3.34s/it, loss=0.0607, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2268/2855 [16:42:59<33:11,  3.39s/it, loss=0.0607, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2268/2855 [16:43:03<33:11,  3.39s/it, loss=0.1288, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2269/2855 [16:43:03<33:00,  3.38s/it, loss=0.1288, lr=2.65e-05]
Epoch 2/8:  79%|███████▉  | 2269/2855 [16:43:06<33:00,  3.38s/it, loss=0.0750, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2270/2855 [16:43:06<32:43,  3.36s/it, loss=0.0750, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2270/2855 [16:43:09<32:43,  3.36s/it, loss=0.1513, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2271/2855 [16:43:09<33:06,  3.40s/it, loss=0.1513, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2271/2855 [16:43:13<33:06,  3.40s/it, loss=0.1972, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2272/2855 [16:43:13<32:35,  3.35s/it, loss=0.1972, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2272/2855 [16:43:16<32:35,  3.35s/it, loss=0.2022, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2273/2855 [16:43:16<32:34,  3.36s/it, loss=0.2022, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2273/2855 [16:43:19<32:34,  3.36s/it, loss=0.1743, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2274/2855 [16:43:19<32:46,  3.38s/it, loss=0.1743, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2274/2855 [16:43:23<32:46,  3.38s/it, loss=0.2900, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2275/2855 [16:43:23<32:32,  3.37s/it, loss=0.2900, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2275/2855 [16:43:26<32:32,  3.37s/it, loss=0.1130, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2276/2855 [16:43:26<32:35,  3.38s/it, loss=0.1130, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2276/2855 [16:43:30<32:35,  3.38s/it, loss=0.2868, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2277/2855 [16:43:30<32:52,  3.41s/it, loss=0.2868, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2277/2855 [16:43:33<32:52,  3.41s/it, loss=0.1597, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2278/2855 [16:43:33<32:47,  3.41s/it, loss=0.1597, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2278/2855 [16:43:36<32:47,  3.41s/it, loss=0.0265, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2279/2855 [16:43:36<32:23,  3.37s/it, loss=0.0265, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2279/2855 [16:43:40<32:23,  3.37s/it, loss=0.1000, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2280/2855 [16:43:40<32:16,  3.37s/it, loss=0.1000, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2280/2855 [16:43:43<32:16,  3.37s/it, loss=0.1382, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2281/2855 [16:43:43<32:25,  3.39s/it, loss=0.1382, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2281/2855 [16:43:46<32:25,  3.39s/it, loss=0.3169, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2282/2855 [16:43:46<32:13,  3.37s/it, loss=0.3169, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2282/2855 [16:43:50<32:13,  3.37s/it, loss=0.2939, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2283/2855 [16:43:50<32:24,  3.40s/it, loss=0.2939, lr=2.65e-05]
Epoch 2/8:  80%|███████▉  | 2283/2855 [16:43:54<32:24,  3.40s/it, loss=0.0948, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2284/2855 [16:43:54<32:57,  3.46s/it, loss=0.0948, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2284/2855 [16:44:02<32:57,  3.46s/it, loss=0.1523, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2285/2855 [16:44:02<47:09,  4.96s/it, loss=0.1523, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2285/2855 [16:44:06<47:09,  4.96s/it, loss=0.1319, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2286/2855 [16:44:06<43:20,  4.57s/it, loss=0.1319, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2286/2855 [16:44:09<43:20,  4.57s/it, loss=0.1671, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2287/2855 [16:44:09<40:07,  4.24s/it, loss=0.1671, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2287/2855 [16:44:12<40:07,  4.24s/it, loss=0.1800, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2288/2855 [16:44:12<37:06,  3.93s/it, loss=0.1800, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2288/2855 [16:44:16<37:06,  3.93s/it, loss=0.2523, lr=2.65e-05]
Epoch 2/8:  80%|████████  | 2289/2855 [16:44:16<36:06,  3.83s/it, loss=0.2523, lr=2.65e-05]
\ No newline at end of file