Link Search Menu Expand Document
enuSpace-tensorflow Manual

QuantizedBatchNormWithGlobalNormalization

tensorflow C++ API

tensorflow::ops::QuantizedBatchNormWithGlobalNormalization

Quantized Batch normalization.

Summary

This op is deprecated and will be removed in the future. Prefertf.nn.batch_normalization.

Arguments:

  • scope: A Scope object
  • t: A 4D input Tensor.
  • t_min: The value represented by the lowest quantized input.
  • t_max: The value represented by the highest quantized input.
  • m: A 1D mean Tensor with size matching the last dimension of t. This is the first output from tf.nn.moments, or a saved moving average thereof.
  • m_min: The value represented by the lowest quantized mean.
  • m_max: The value represented by the highest quantized mean.
  • v: A 1D variance Tensor with size matching the last dimension of t. This is the second output from tf.nn.moments, or a saved moving average thereof.
  • v_min: The value represented by the lowest quantized variance.
  • v_max: The value represented by the highest quantized variance.
  • beta: A 1D beta Tensor with size matching the last dimension of t. An offset to be added to the normalized tensor.
  • beta_min: The value represented by the lowest quantized offset.
  • beta_max: The value represented by the highest quantized offset.
  • gamma: A 1D gamma Tensor with size matching the last dimension of t. If “scale_after_normalization” is true, this tensor will be multiplied with the normalized tensor.
  • gamma_min: The value represented by the lowest quantized gamma.
  • gamma_max: The value represented by the highest quantized gamma.
  • variance_epsilon: A small float number to avoid dividing by 0.
  • scale_after_normalization: A bool indicating whether the resulted tensor needs to be multiplied with gamma.

Returns:

QuantizedAvgPool block

Source link : https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_nn.cpp

Argument:

  • Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.)
  • Input t: connect Input node.
  • Input t_min: connect Input node.
  • Input t_max: connect Input node.
  • Input m: connect Input node.
  • Input m_min: connect Input node.
  • Input m_max: connect Input node.
  • Input v: connect Input node.
  • Input v_min: connect Input node.
  • Input v_max: connect Input node.
  • Input beta: connect Input node.
  • Input beta_min: connect Input node.
  • Input beta_max: connect Input node.
  • Input gamma: connect Input node.
  • Input gamma_min: connect Input node.
  • Input gamma_max: connect Input node.
  • DataType out_type : input DataType in value.
  • float variance_epsilon: input float in value.
  • bool scale_after_normalization: input bool in value.

Return:

  • Output result: Output object of QuantizedAvgPool class object.
  • Output result_min: Output object of QuantizedAvgPool class object.
  • Output result_max: Output object of QuantizedAvgPool class object.

Result:

  • std::vector(Tensor) result_result : Returned object of executed result by calling session.
  • std::vector(Tensor) result_result_min : Returned object of executed result by calling session.
  • std::vector(Tensor) result_result_max : Returned object of executed result by calling session.

Using Method