Class BatchNorm

Defined in File batch_norm.h

Inheritance Relationships

Base Type

public mindspore::ops::BaseOperator

Derived Types

public mindspore::ops::BatchNormWithActivation (Class BatchNormWithActivation)
public mindspore::ops::BatchNormWithAddAndActivation (Class BatchNormWithAddAndActivation)

Class Documentation

class BatchNorm : public mindspore::ops::BaseOperator 

Batch Normalization for input data and updated parameters. Refer to Python API mindspore.ops.BatchNorm for more details.

Subclassed by mindspore::ops::BatchNormWithActivation, mindspore::ops::BatchNormWithAddAndActivation

Public Functions

inline BatchNorm(): Constructor.

void Init(const bool is_training = false, const float epsilon = 1e-5, const float momentun = 0.1, const Format &format = NCHW): Init. Refer to the parameters of Python API mindspore.ops.BatchNorm for the inputs.

void set_is_training(const bool is_training): Set is_training.

void set_epsilon(const float epsilon): Set epsilon.

void set_format(const Format &format): Set format.

void set_momentum(const float momentum): Set momentum.

bool get_is_training() const

Get is_training.

返回: is_training.

float get_epsilon() const

Get epsilon.

返回: epsilon.

Format get_format() const

Get format.

返回: format.

float get_momentum() const

Get momentum.

返回: momentum.