Shortcuts

LambdaCallback

class pytorch_lightning.callbacks.LambdaCallback(on_before_accelerator_backend_setup=None, setup=None, on_configure_sharded_model=None, teardown=None, on_init_start=None, on_init_end=None, on_fit_start=None, on_fit_end=None, on_sanity_check_start=None, on_sanity_check_end=None, on_train_batch_start=None, on_train_batch_end=None, on_train_epoch_start=None, on_train_epoch_end=None, on_validation_epoch_start=None, on_validation_epoch_end=None, on_test_epoch_start=None, on_test_epoch_end=None, on_epoch_start=None, on_epoch_end=None, on_batch_start=None, on_validation_batch_start=None, on_validation_batch_end=None, on_test_batch_start=None, on_test_batch_end=None, on_batch_end=None, on_train_start=None, on_train_end=None, on_pretrain_routine_start=None, on_pretrain_routine_end=None, on_validation_start=None, on_validation_end=None, on_test_start=None, on_test_end=None, on_keyboard_interrupt=None, on_exception=None, on_save_checkpoint=None, on_load_checkpoint=None, on_before_backward=None, on_after_backward=None, on_before_optimizer_step=None, on_before_zero_grad=None, on_predict_start=None, on_predict_end=None, on_predict_batch_start=None, on_predict_batch_end=None, on_predict_epoch_start=None, on_predict_epoch_end=None)[source]

Bases: pytorch_lightning.callbacks.base.Callback

Create a simple callback on the fly using lambda functions.

Parameters

**kwargs – hooks supported by Callback

Example:

>>> from pytorch_lightning import Trainer
>>> from pytorch_lightning.callbacks import LambdaCallback
>>> trainer = Trainer(callbacks=[LambdaCallback(setup=lambda *args: print('setup'))])