tensorflow::ops::LRN::Attrs

#include <nn_ops.h>

Optional attribute setters for LRN.

Summary

Public functions
`Alpha(float x)`	`TF_MUST_USE_RESULT Attrs` A scale factor, usually positive.
`Beta(float x)`	`TF_MUST_USE_RESULT Attrs` An exponent.
`Bias(float x)`	`TF_MUST_USE_RESULT Attrs` An offset (usually positive to avoid dividing by 0).
`DepthRadius(int64 x)`	`TF_MUST_USE_RESULT Attrs` 0-D.

float tensorflow::ops::LRN::Attrs::alpha_ = 1.0f

float tensorflow::ops::LRN::Attrs::beta_ = 0.5f

float tensorflow::ops::LRN::Attrs::bias_ = 1.0f

int64 tensorflow::ops::LRN::Attrs::depth_radius_ = 5

TF_MUST_USE_RESULT Attrs tensorflow::ops::LRN::Attrs::Alpha(
  float x
)

A scale factor, usually positive.

Defaults to 1

TF_MUST_USE_RESULT Attrs tensorflow::ops::LRN::Attrs::Beta(
  float x
)

An exponent.

Defaults to 0.5

TF_MUST_USE_RESULT Attrs tensorflow::ops::LRN::Attrs::Bias(
  float x
)

An offset (usually positive to avoid dividing by 0).

Defaults to 1

TF_MUST_USE_RESULT Attrs tensorflow::ops::LRN::Attrs::DepthRadius(
  int64 x
)

0-D.

Half-width of the 1-D normalization window.

Defaults to 5

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.

Last updated 2021-11-15 UTC.