Package ai.djl.nn.norm

Class BatchNorm.Builder

java.lang.Object

ai.djl.nn.norm.BatchNorm.Builder

Enclosing class:

BatchNorm

public static final class BatchNorm.Builder
extends java.lang.Object

The Builder to construct a BatchNorm.

Method Summary

Modifier and Type	Method	Description
BatchNorm	build()	Builds a BatchNorm block.
BatchNorm.Builder	optAxis(int axis)	Set the axis in which channel is specified.
BatchNorm.Builder	optCenter(boolean val)	If True, add offset of `beta` to normalized tensor.
BatchNorm.Builder	optEpsilon(float val)	Sets the epsilon value to prevent division by 0.
BatchNorm.Builder	optMomentum(float val)	Set the momentum for moving average.
BatchNorm.Builder	optScale(boolean val)	If True, multiply result by `gamma`.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

optAxis

public BatchNorm.Builder optAxis(int axis)

Set the axis in which channel is specified. Defaults to 1.

Parameters:

axis - the axis in which channel is specified

Returns:

this Builder

optCenter

public BatchNorm.Builder optCenter(boolean val)

If True, add offset of `beta` to normalized tensor. Defaults to True.

Parameters:

val - True or False on whether to add and train offset value

Returns:

this Builder

optScale

public BatchNorm.Builder optScale(boolean val)

If True, multiply result by `gamma`. Defaults to True;

Parameters:

val - True or False on whether to add and train scale value

Returns:

this Builder

optEpsilon

public BatchNorm.Builder optEpsilon(float val)

Sets the epsilon value to prevent division by 0.

Parameters:

val - the epsilon value

Returns:

this Builder

optMomentum

public BatchNorm.Builder optMomentum(float val)

Set the momentum for moving average.

Parameters:

val - the momentum for moving average

Returns:

this Builder

build

public BatchNorm build()

Builds a BatchNorm block.

Returns:

the BatchNorm block