TensorFlow 2.0 Beta is available Learn more

tensorflow::ops::CropAndResize::Attrs

#include <image_ops.h>

Optional attribute setters for CropAndResize.

Summary

Public attributes

extrapolation_value_ = 0.0f
float
method_ = "bilinear"
StringPiece

Public functions

ExtrapolationValue(float x)
TF_MUST_USE_RESULT Attrs
Value used for extrapolation, when applicable.
Method(StringPiece x)
TF_MUST_USE_RESULT Attrs
A string specifying the sampling method for resizing.

Public attributes

extrapolation_value_

float tensorflow::ops::CropAndResize::Attrs::extrapolation_value_ = 0.0f

method_

StringPiece tensorflow::ops::CropAndResize::Attrs::method_ = "bilinear"

Public functions

ExtrapolationValue

TF_MUST_USE_RESULT Attrs tensorflow::ops::CropAndResize::Attrs::ExtrapolationValue(
  float x
)

Value used for extrapolation, when applicable.

Defaults to 0

Method

TF_MUST_USE_RESULT Attrs tensorflow::ops::CropAndResize::Attrs::Method(
  StringPiece x
)

A string specifying the sampling method for resizing.

It can be either "bilinear" or "nearest" and default to "bilinear". Currently two sampling methods are supported: Bilinear and Nearest Neighbor.

Defaults to "bilinear"