Translation with marianmt. Early stopping stucked

[John6666]

If no error message appears, it means that the processing has completed normally (although the result may not be normal), so there may be something wrong with the evaluation. The following example shows an error message, so it is slightly different, but it is similar to this…

However, even if a callback is set so that the early stopping condition is always true, it seems that it will be executed for one epoch…

Early stopping callback problem

Hello, I am having problems with the EarlyStoppingCallback I set up in my trainer class as below: training_args = TrainingArguments( output_dir = 'BERT', num_train_epochs = epochs, do_train = True, do_eval = True, evaluation_strategy = 'epoch', logging_strategy = 'epoch', per_device_train_batch_size = batch_size, per_device_eval_batch_size = batch_size, warmup_steps = 250, weight_decay = 0.01, fp16 = True, metric_for_best_model = 'eval_loss', …

https://stackoverflow.com/questions/76330546/how-to-determine-the-value-of-early-stopping-patience-in-huggingfaces-seq2seqtr

[Maybe Bug] When using EarlyStopping Callbacks with Seq2SeqTraininer, training didn't stop

When trying to use EarlyStopping for Seq2SeqTrainer, e.g. patience was set to 1 and threshold 1.0: training_args = Seq2SeqTrainingArguments( output_dir='./', num_train_epochs=3, per_device_train_batch_size=4, per_device_eval_batch_size=4, logging_steps=1, save_steps=5, eval_steps=1, max_steps=10, evaluation_strategy="steps", predict_with_generate=True, report_to=None, metric_for_best_model="chr_f_score", load_best_model_at_end=True ) early_st…