Module: tf.keras  |  TensorFlow v2.16.1 (original) (raw)

Module: tf.keras

Stay organized with collections Save and categorize content based on your preferences.

DO NOT EDIT.

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

Modules

activations module: DO NOT EDIT.

applications module: DO NOT EDIT.

backend module: DO NOT EDIT.

callbacks module: DO NOT EDIT.

config module: DO NOT EDIT.

constraints module: DO NOT EDIT.

datasets module: DO NOT EDIT.

distribution module: DO NOT EDIT.

dtype_policies module: DO NOT EDIT.

export module: DO NOT EDIT.

initializers module: DO NOT EDIT.

layers module: DO NOT EDIT.

legacy module: DO NOT EDIT.

losses module: DO NOT EDIT.

metrics module: DO NOT EDIT.

mixed_precision module: DO NOT EDIT.

models module: DO NOT EDIT.

ops module: DO NOT EDIT.

optimizers module: DO NOT EDIT.

preprocessing module: DO NOT EDIT.

quantizers module: DO NOT EDIT.

random module: DO NOT EDIT.

regularizers module: DO NOT EDIT.

tree module: DO NOT EDIT.

utils module: DO NOT EDIT.

Classes

class DTypePolicy: A dtype policy for a Keras layer.

class FloatDTypePolicy: A dtype policy for a Keras layer.

class Function: Class that encapsulates a computation graph of Keras operations.

class Initializer: Initializer base class: all Keras initializers inherit from this class.

class InputSpec: Specifies the rank, dtype and shape of every input to a layer.

class KerasTensor: Symbolic tensor -- encapsulates a shape and a dtype.

class Layer: This is the class from which all layers inherit.

class Loss: Loss base class.

class Metric: Encapsulates metric logic and state.

class Model: A model grouping layers into an object with training/inference features.

class Operation

class Optimizer: A class for Tensorflow specific optimizer logic.

class Quantizer

class Regularizer: Regularizer base class.

class Sequential: Sequential groups a linear stack of layers into a Model.

class StatelessScope: Scope to prevent any update to Keras Variables.

class Variable: Represents a backend-agnostic variable in Keras.

class name_scope: Creates a sub-namespace for variable paths.

Functions

Input(...): Used to instantiate a Keras tensor.

device(...)

version(...)