MultiTrainingCallback

class MultiTrainingCallback(callbacks=None, callback_kwargs=None)

Bases: pykeen.training.callbacks.TrainingCallback

A wrapper for calling multiple training callbacks together.

Initialize the callback.

Note

the constructor allows “broadcasting” of callbacks, i.e., proving a single callback, but a list of callback kwargs. In this case, for each element of this list the given callback is instantiated.

Parameters

• callbacks (Union[str, TrainingCallback, Type[TrainingCallback], None, Sequence[Union[str, TrainingCallback, Type[TrainingCallback], None]]]) – the callbacks

• callback_kwargs (Union[Mapping[str, Any], None, Sequence[Optional[Mapping[str, Any]]]]) – additional keyword-based parameters for instantiating the callbacks

Methods Summary

on_batch(epoch, batch, batch_loss, **kwargs)	Call for training batches.
post_batch(epoch, batch, **kwargs)	Call for training batches.
post_epoch(epoch, epoch_loss, **kwargs)	Call after epoch.
post_train(losses, **kwargs)	Call after training.
pre_step(**kwargs)	Call before the optimizer's step.
register_callback(callback)	Register a callback.
register_training_loop(loop)	Register the training loop.

Methods Documentation

on_batch(epoch, batch, batch_loss, **kwargs)

Call for training batches.

Return type

None

post_batch(epoch, batch, **kwargs)

Call for training batches.

Return type

None

post_epoch(epoch, epoch_loss, **kwargs)

Call after epoch.

Return type

None

post_train(losses, **kwargs)

Call after training.

Return type

None

pre_step(**kwargs)

Call before the optimizer’s step.

Return type

None

register_callback(callback)

Register a callback.

Return type

None

register_training_loop(loop)

Register the training loop.

Return type

None