API ReferenceΒΆ

ashpy.ashtypes Custom Type-Aliases.
ashpy.callbacks Callbacks in order to gain control over the training loop.
ashpy.contexts Contexts help gaining an easier control over the model selection and testing process of the models.
ashpy.keras Custom extensions of standard Keras components.
ashpy.layers Collection of layers.
ashpy.losses Collection of Losses.
ashpy.metrics Collection of Metrics.
ashpy.models Collection of Models.
ashpy.modes Various modalities used to configure certain ash behaviours.
ashpy.restorers Restorers allow for easy restoration of tracked objects from tf.train.Checkpoint.
ashpy.trainers Trainers help reducing boilerplate code by bootstrapping models training.