trw.train.callback_export_samples

Module Contents

Classes

CallbackExportSamples

Defines a callback function that may be called before training, during training, after training

Functions

callbacks_per_loss_term(dataset_name, split_name, batch, loss_terms, root, datasets_infos, loss_terms_inclusion, feature_exclusions, dataset_exclusions, split_exclusions, exported_cases, max_samples, epoch)

Attributes

logger

trw.train.callback_export_samples.logger

trw.train.callback_export_samples.callbacks_per_loss_term(dataset_name, split_name, batch, loss_terms, root, datasets_infos, loss_terms_inclusion, feature_exclusions, dataset_exclusions, split_exclusions, exported_cases, max_samples, epoch)

class trw.train.callback_export_samples.CallbackExportSamples(max_samples=20, dirname='samples', loss_terms_inclusion=None, feature_exclusions=None, dataset_exclusions=None, split_exclusions=None)

Bases: trw.train.callback.Callback

Defines a callback function that may be called before training, during training, after training

__call__(self, options, history, model, losses, outputs, datasets, datasets_infos, callbacks_per_batch, **kwargs)