Module: tf.keras.distribution

DO NOT EDIT.

This file was autogenerated. Do not edit it by hand, since your modifications would be overwritten.

Classes

class DataParallel: Distribution for data parallelism.

class DeviceMesh: A cluster of computation devices for distributed computation.

class LayoutMap: A dict-like object that maps string to TensorLayout instances.

class ModelParallel: Distribution that shards model variables.

class TensorLayout: A layout to apply to a tensor.

Functions

distribute_tensor(...): Change the layout of a Tensor value in the jit function execution.

distribution(...): Retrieve the current distribution from global context.

initialize(...): Initialize the distribution system for multi-host/process setting.

list_devices(...): Return all the available devices based on the device type.

set_distribution(...): Set the distribution as the global distribution setting.