composer.algorithms.channels_last.channels_last

composer.algorithms.channels_last.channels_last

Functions

apply_channels_last

Changes the memory format of the model to torch.channels_last.

Classes

Algorithm	Base class for algorithms.
ChannelsLast	Changes the memory format of the model to torch.channels_last.
Event	Enum to represent events in the training loop.
Logger	Logger routes metrics to the LoggerCallback.
State	The state of the trainer.

Attributes

• Optional

• annotations

• log

class composer.algorithms.channels_last.channels_last.ChannelsLast

Bases: composer.core.algorithm.Algorithm

Changes the memory format of the model to torch.channels_last. This usually yields improved GPU utilization.

Runs on Event.INIT, so it can set the memory format before the model is DDP wrapped. Has no hyperparameters.

apply(event, state, logger)

Changes the memory format of the model to torch.channels_last

match(event, state)

Runs on Event.INIT

composer.algorithms.channels_last.channels_last.apply_channels_last(model)

Changes the memory format of the model to torch.channels_last.

This usually yields improved GPU utilization.

Parameters

model – model or module to modify