Error resuming training of a NER model

[fmafelipe]

I want to resume training a NER model as shown in the tutorials, loading the model checkpoint but running it with :

trainer.resume(trained_model, base_path=path + '-resume', max_epochs=25, )

It simply shows me the metrics of the loaded model and does not perform any training.

[helpmefindaname]

hi @fmafelipe ,
setting max_epochs=25 means that the model will be trained for a total of 25 epochs, not for additional 25 epochs.
if you have already trained the model for 25 epochs, then the model will be trained for 0 epochs (e.g. not at all).

[fmafelipe]

Hi @helpmefindaname
So following this example, if I want to train the model for an additional 25 epochs, do I need to set max_epochs=50 ?
Thanks for the help

[helpmefindaname]

assuming that you trained for 25 epochs, yes.

If you are using the latest master branch, (where #2895 is merged) you can also use additional_epochs=25 instead.

[fmafelipe]

Thank you so much 👍👌

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.