Callbacks
base
¶
Callback (ABC)
¶
Callback objects define events on which it will run during the model training cycle.
clean(self)
¶
Clean up
on_epoch_cancel(self)
¶
Called after epoch is cancelled
on_epoch_end(self)
¶
Called after each epoch
on_epoch_start(self)
¶
Called Before each Epoch
on_fit_cancel(self)
¶
Called after model.fit(...)
is cancelled
on_fit_end(self)
¶
Called after model.fit(...)
on_fit_start(self)
¶
Called on each model.fit(...)
on_forward_cancel(self)
¶
Called after model.forward(...) is cancelled
on_forward_end(self)
¶
Called after model.forward(...)
on_forward_start(self)
¶
Called before model.forward(...)
on_train_epoch_cancel(self)
¶
Called after training epoch is cancelled
on_train_epoch_end(self, *args, **kwargs)
¶
Called after end of training epoch
on_train_epoch_start(self)
¶
Called on start of training epoch
on_train_step_cancel(self)
¶
Called after training step is cancelled
on_train_step_end(self, *args, **kwargs)
¶
Called after training step
on_train_step_start(self)
¶
called before train_step
on_val_epoch_cancel(self)
¶
called after validation epoch cancelled
on_val_epoch_end(self, *args, **kwargs)
¶
called after validation epoch ends
on_val_epoch_start(self)
¶
Called on start of validation epoch
on_val_step_cancel(self)
¶
Called after validation step is cancelled
on_val_step_end(self, *args, **kwargs)
¶
Called after validation step
on_val_step_start(self)
¶
Called on validation step
with_event(self, event_type, func, exception, final_fn=<function dummy at 0x7f6a30285a60>)
¶
Calls a function with event wrapped around. Inspired from FastAI. Ref: https://github.com/fastai/fastai/blob/6e44b354f4d12bdfa2c9530f38f851c54a05764d/fastai/learner.py#L162
comet
¶
CometCallback (Callback)
¶
Comet Logging callback. To use this callback pip install comet-ml
.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
project_name |
str |
Name of the Project |
'awesome-project' |
api_key |
Optional[str] |
project API key |
None |
offline |
bool |
log experiment offline |
False |
on_epoch_end(self)
¶
Called after each epoch
on_fit_start(self)
¶
Called on each model.fit(...)
on_train_epoch_start(self)
¶
Called on start of training epoch
on_train_step_end(self, outputs=None, **_)
¶
Called after training step
on_val_epoch_start(self)
¶
Called on start of validation epoch
on_val_step_end(self, outputs=None, **_)
¶
Called after validation step
gpu
¶
EmissionTrackerCallback (Callback)
¶
Tracks the carbon emissions produced by deep neural networks using
CodeCarbon. To use this callback first install codecarbon using
pip install codecarbon
.
For offline use, you must have to specify the country code.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
offline |
bool |
whether to use internet connection or not. You will have to provide the country code |
False |
**kwargs |
passed directly to codecarbon class. |
{} |
on_fit_end(self)
¶
Called after model.fit(...)
logger
¶
CSVLogger (Callback)
¶
Saves Model training metrics as CSV
Parameters:
Name | Type | Description | Default |
---|---|---|---|
filename |
str |
filename of the csv |
'./experiment.csv' |
path |
str |
folder path location of the csv |
'/home/docs/checkouts/readthedocs.org/user_builds/gradsflow/checkouts/v0.0.8' |
verbose |
bool |
Whether to show output |
False |
on_epoch_end(self)
¶
Called after each epoch
ModelCheckpoint (Callback)
¶
Saves Model checkpoint
Parameters:
Name | Type | Description | Default |
---|---|---|---|
filename |
Optional[str] |
name of checkpoint |
None |
path |
str |
folder path location of the model checkpoint. Will create a folder if does not exist. |
'/home/docs/checkouts/readthedocs.org/user_builds/gradsflow/checkouts/v0.0.8' |
save_extra |
bool |
whether to save extra details like tracker |
False |
on_epoch_end(self)
¶
Called after each epoch
progress
¶
ProgressCallback (Callback)
¶
clean(self)
¶
Clean up
on_epoch_end(self)
¶
Called after each epoch
on_fit_end(self)
¶
Called after model.fit(...)
on_fit_start(self)
¶
Called on each model.fit(...)
on_train_epoch_end(self, *args, **kwargs)
¶
Called after end of training epoch
on_train_epoch_start(self)
¶
Called on start of training epoch
on_train_step_end(self, *args, **kwargs)
¶
Called after training step
on_val_epoch_end(self, *args, **kwargs)
¶
called after validation epoch ends
on_val_epoch_start(self)
¶
Called on start of validation epoch
on_val_step_end(self, *args, **kwargs)
¶
Called after validation step
raytune
¶
runner
¶
CallbackRunner (Callback)
¶
clean(self, keep=None)
¶
Remove all the callbacks except callback names provided in keep
on_epoch_end(self)
¶
Called after each epoch
on_epoch_start(self)
¶
Called Before each Epoch
on_fit_end(self)
¶
Called after model.fit(...)
on_fit_start(self)
¶
Called on each model.fit(...)
on_forward_end(self)
¶
Called after model.forward(...)
on_forward_start(self)
¶
Called before model.forward(...)
on_train_epoch_end(self, *args, **kwargs)
¶
Called after end of training epoch
on_train_epoch_start(self)
¶
Called on start of training epoch
on_train_step_end(self, *args, **kwargs)
¶
Called after training step
on_train_step_start(self)
¶
called before train_step
on_val_epoch_end(self, *args, **kwargs)
¶
called after validation epoch ends
on_val_epoch_start(self)
¶
Called on start of validation epoch
on_val_step_end(self, *args, **kwargs)
¶
Called after validation step
on_val_step_start(self)
¶
Called on validation step
training
¶
TrainEvalCallback (Callback)
¶
on_train_epoch_start(self)
¶
Called on start of training epoch
on_train_step_end(self, *args, *, outputs=None, **kwargs)
¶
Called after training step
on_train_step_start(self)
¶
called before train_step
on_val_epoch_start(self)
¶
Called on start of validation epoch
on_val_step_end(self, *args, *, outputs=None, **kwargs)
¶
Called after validation step
wandb
¶
WandbCallback (Callback)
¶
Weights & Biases Logging callback. To use this callback pip install wandb
.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
log_model |
bool |
Whether to upload model artifact to Wandb |
False |
code_file |
Optional[str] |
path of the code you want to upload as artifact to Wandb |
None |