the thumt transformer model no implement early stopping to prevent overfitting? about thumt HOT 1 CLOSED

Early stopping can be achieved in two ways in THUMT:

1. When --validation <your validation source file> --references <your validation reference files> is set in training, THUMT will evaluate on the validation set periodically. The checkpoints that achieves highest BLEU scores on the validation set will be saved in the train/eval folder. (You can halt or resume the training whenever you want.)
2. Setting keep_checkpoint_max in parameters to reserve as many checkpoints as you want.

from thumt.