config.json

{
  "activation_function": "relu",
  "architectures": [
    "Mask2FormerForUniversalSegmentation"
  ],
  "backbone": null,
  "backbone_config": {
    "depths": [
      3,
      4,
      6,
      3
    ],
    "downsample_in_bottleneck": false,
    "downsample_in_first_stage": false,
    "embedding_size": 64,
    "hidden_act": "relu",
    "hidden_sizes": [
      256,
      512,
      1024,
      2048
    ],
    "layer_type": "bottleneck",
    "model_type": "resnet",
    "num_channels": 3,
    "out_features": [
      "stage1",
      "stage2",
      "stage3",
      "stage4"
    ],
    "out_indices": [
      1,
      2,
      3,
      4
    ],
    "stage_names": [
      "stem",
      "stage1",
      "stage2",
      "stage3",
      "stage4"
    ]
  },
  "backbone_kwargs": null,
  "class_weight": 2.0,
  "common_stride": 4,
  "decoder_layers": 10,
  "dice_weight": 5.0,
  "dim_feedforward": 2048,
  "dropout": 0.0,
  "dtype": "float32",
  "encoder_feedforward_dim": 1024,
  "encoder_layers": 6,
  "enforce_input_projection": false,
  "feature_size": 256,
  "feature_strides": [
    4,
    8,
    16,
    32
  ],
  "hidden_dim": 256,
  "id2label": {
    "0": "Thing"
  },
  "ignore_value": 255,
  "importance_sample_ratio": 0.75,
  "init_std": 0.02,
  "init_xavier_std": 1.0,
  "label2id": {
    "Thing": 0
  },
  "mask_feature_size": 256,
  "mask_weight": 5.0,
  "model_type": "mask2former",
  "no_object_weight": 0.1,
  "num_attention_heads": 8,
  "num_hidden_layers": 10,
  "num_queries": 2000,
  "output_auxiliary_logits": null,
  "oversample_ratio": 3.0,
  "pre_norm": false,
  "train_num_points": 12544,
  "transformers_version": "4.57.1",
  "use_auxiliary_loss": true,
  "use_pretrained_backbone": false,
  "use_timm_backbone": false
}