Callbacks that make decisions depending how a monitored metric/loss behaves

class TerminateOnNaNCallback[source]

TerminateOnNaNCallback(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_backward=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) :: Callback

A Callback that terminates training if loss is NaN.

learn = synth_learner()
learn.fit(10, lr=100, cbs=TerminateOnNaNCallback())
epoch train_loss valid_loss time
0 215824033216875417685323736416256.000000 00:00
assert len(learn.recorder.losses) < 10 * len(learn.dls.train)
for l in learn.recorder.losses:
    assert not torch.isinf(l) and not torch.isnan(l) 

class TrackerCallback[source]

TrackerCallback(monitor='valid_loss', comp=None, min_delta=0.0, reset_on_fit=True) :: Callback

A Callback that keeps track of the best value in monitor.

When implementing a Callback that has behavior that depends on the best value of a metric or loss, subclass this Callback and use its best (for best value so far) and new_best (there was a new best value this epoch) attributes. If you want to maintain best over subsequent calls to fit (e.g., Learner.fit_one_cycle), set reset_on_fit = True.

comp is the comparison operator used to determine if a value is best than another (defaults to np.less if 'loss' is in the name passed in monitor, np.greater otherwise) and min_delta is an optional float that requires a new value to go over the current best (depending on comp) by at least that amount.

class EarlyStoppingCallback[source]

EarlyStoppingCallback(monitor='valid_loss', comp=None, min_delta=0.0, patience=1, reset_on_fit=True) :: TrackerCallback

A TrackerCallback that terminates training when monitored quantity stops improving.

comp is the comparison operator used to determine if a value is best than another (defaults to np.less if 'loss' is in the name passed in monitor, np.greater otherwise) and min_delta is an optional float that requires a new value to go over the current best (depending on comp) by at least that amount. patience is the number of epochs you're willing to wait without improvement.

learn = synth_learner(n_trn=2, metrics=F.mse_loss)
learn.fit(n_epoch=200, lr=1e-7, cbs=EarlyStoppingCallback(monitor='mse_loss', min_delta=0.1, patience=2))
epoch train_loss valid_loss mse_loss time
0 7.289688 6.762055 6.762055 00:00
1 7.300182 6.762038 6.762038 00:00
2 7.304251 6.762012 6.762012 00:00
No improvement since epoch 0: early stopping
learn.validate()
(#2) [6.762012481689453,6.762012481689453]
learn = synth_learner(n_trn=2)
learn.fit(n_epoch=200, lr=1e-7, cbs=EarlyStoppingCallback(monitor='valid_loss', min_delta=0.1, patience=2))
epoch train_loss valid_loss time
0 17.360760 18.424431 00:00
1 17.410801 18.424389 00:00
2 17.400934 18.424322 00:00
No improvement since epoch 0: early stopping

class SaveModelCallback[source]

SaveModelCallback(monitor='valid_loss', comp=None, min_delta=0.0, fname='model', every_epoch=False, with_opt=False, reset_on_fit=True) :: TrackerCallback

A TrackerCallback that saves the model's best during training and loads it at the end.

comp is the comparison operator used to determine if a value is best than another (defaults to np.less if 'loss' is in the name passed in monitor, np.greater otherwise) and min_delta is an optional float that requires a new value to go over the current best (depending on comp) by at least that amount. Model will be saved in learn.path/learn.model_dir/name.pth, maybe every_epoch or at each improvement of the monitored quantity.

learn = synth_learner(n_trn=2, path=Path.cwd()/'tmp')
learn.fit(n_epoch=2, cbs=SaveModelCallback())
assert (Path.cwd()/'tmp/models/model.pth').exists()
learn.fit(n_epoch=2, cbs=SaveModelCallback(every_epoch=True))
for i in range(2): assert (Path.cwd()/f'tmp/models/model_{i}.pth').exists()
shutil.rmtree(Path.cwd()/'tmp')
epoch train_loss valid_loss time
0 16.789867 15.243573 00:00
1 16.694817 14.915352 00:00
Better model found at epoch 0 with valid_loss value: 15.243573188781738.
Better model found at epoch 1 with valid_loss value: 14.915351867675781.
epoch train_loss valid_loss time
0 16.171261 14.440922 00:00
1 15.897676 13.870268 00:00

ReduceLROnPlateau

class ReduceLROnPlateau[source]

ReduceLROnPlateau(monitor='valid_loss', comp=None, min_delta=0.0, patience=1, factor=10.0, min_lr=0, reset_on_fit=True) :: TrackerCallback

A TrackerCallback that reduces learning rate when a metric has stopped improving.

learn = synth_learner(n_trn=2)
learn.fit(n_epoch=4, lr=1e-7, cbs=ReduceLROnPlateau(monitor='valid_loss', min_delta=0.1, patience=2))
epoch train_loss valid_loss time
0 25.698519 40.559010 00:00
1 25.731672 40.558941 00:00
2 25.716520 40.558846 00:00
3 25.729078 40.558838 00:00
Epoch 2: reducing lr to 1e-08
learn = synth_learner(n_trn=2)
learn.fit(n_epoch=6, lr=5e-8, cbs=ReduceLROnPlateau(monitor='valid_loss', min_delta=0.1, patience=2, min_lr=1e-8))
epoch train_loss valid_loss time
0 6.026482 5.962046 00:00
1 6.027329 5.962038 00:00
2 6.024285 5.962028 00:00
3 6.024290 5.962025 00:00
4 6.022497 5.962022 00:00
5 6.023217 5.962019 00:00
Epoch 2: reducing lr to 1e-08