Tracking Callbacks

Additional callbacks which make decisions depending how a monitored metric/loss behaves.

source

TerminateOnTrainNaN

 TerminateOnTrainNaN (after_create=None, before_fit=None,
                      before_epoch=None, before_train=None,
                      before_batch=None, after_pred=None, after_loss=None,
                      before_backward=None, after_cancel_backward=None,
                      after_backward=None, before_step=None,
                      after_cancel_step=None, after_step=None,
                      after_cancel_batch=None, after_batch=None,
                      after_cancel_train=None, after_train=None,
                      before_validate=None, after_cancel_validate=None,
                      after_validate=None, after_cancel_epoch=None,
                      after_epoch=None, after_cancel_fit=None,
                      after_fit=None)

A Callback that terminates training if the training loss is NaN and ignores valid loss.


source

SaveModelAtEnd

 SaveModelAtEnd (fname='model', with_opt=False)

A SaveModelCallback which only saves the model at the end so loggers can find it.

Type Default Details
fname str model Model filename
with_opt bool False Include optimizer state

source

LastMetricCallback

 LastMetricCallback (metrics:Listified[str]|None=None)

A Callback which stores the last metric(s) value by name (or all if None) in the Learner.lastmetric dictionary