tensorflow::ops::QuantizedInstanceNorm

#include <array_ops.h>

Quantized Instance normalization.

Summary

Arguments:

• scope: A Scope object
• x: A 4D input Tensor.
• x_min: The value represented by the lowest quantized input.
• x_max: The value represented by the highest quantized input.

Optional attributes (see Attrs):

• output_range_given: If True, given_y_min and given_y_min and given_y_max are used as the output range. Otherwise, the implementation computes the output range.
• given_y_min: Output in y_min if output_range_given is True.
• given_y_max: Output in y_max if output_range_given is True.
• variance_epsilon: A small float number to avoid dividing by 0.
• min_separation: Minimum value of y_max - y_min

Returns:

• Output y: A 4D Tensor.
• Output y_min: The value represented by the lowest quantized output.
• Output y_max: The value represented by the highest quantized output.

Public attributes

::tensorflow::Output y

::tensorflow::Output y_max

::tensorflow::Output y_min

Public functions

 QuantizedInstanceNorm(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input x,
  ::tensorflow::Input x_min,
  ::tensorflow::Input x_max
)

 QuantizedInstanceNorm(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input x,
  ::tensorflow::Input x_min,
  ::tensorflow::Input x_max,
  const QuantizedInstanceNorm::Attrs & attrs
)

Public static functions

Attrs GivenYMax(
  float x
)

Attrs GivenYMin(
  float x
)

Attrs MinSeparation(
  float x
)

Attrs OutputRangeGiven(
  bool x
)

Attrs VarianceEpsilon(
  float x
)