finetuner.tuner.callback.best_model_checkpoint module#

class finetuner.tuner.callback.best_model_checkpoint.BestModelCheckpoint(save_dir, monitor='val_loss', mode='auto', verbose=False)[source]#

Bases: finetuner.tuner.callback.base.BaseCallback

Callback to save the best model across all epochs

An option this callback provides include: - Definition of ‘best’; which quantity to monitor and whether it should be

maximized or minimized.

Parameters

save_dir (str) – string, path to save the model file.
monitor (str) – if monitor=’train_loss’ best model saved will be according to the training loss, while if monitor=’val_loss’ best model saved will be according to the validation loss. If monitor is set to an evaluation metric, best model saved will be according to this metric.
mode (str) – one of {‘auto’, ‘min’, ‘max’}. The decision to overwrite the currently saved model is made based on either the maximization or the minimization of the monitored quantity. For an evaluation metric, this should be max, for val_loss this should be min, etc. In auto mode, the mode is set to min if monitor=’loss’ or monitor=’val_loss’ and to min otherwise.
verbose (bool) – Whether to log notifications when a checkpoint is saved.

on_train_batch_end(tuner)[source]#: Called at the end of a training batch, after the backward pass.

on_val_batch_end(tuner)[source]#: Called at the end of an evaluation batch.

on_epoch_end(tuner)[source]#: Called at the end of an epoch, after both training and validation (or just training if no validaton is provided).

static load(tuner, fp)[source]#: Loads the model.