Performs a resize and padding as a preprocess during a convolution.
It's often possible to do spatial transformations more efficiently as part of the packing stage of a convolution, so this op allows for an optimized implementation where these stages are fused together. This prevents the need to write out the intermediate results as whole tensors, reducing memory pressure, and we can get some latency gains by merging the transformation calculations. The data_format attribute for Conv2D isn't supported by this op, and defaults to 'NHWC' order. Internally this op uses a single per-graph scratch buffer, which means that it will block if multiple versions are being run in parallel. This is because this operator is primarily an optimization to minimize memory usage.
[batch, in_height, in_width, in_channels].
new_height, new_width. The new size for the images.
[filter_height, filter_width, in_channels, out_channels].
input. Must be in the same order as the dimension specified with format.
Optional attributes (see
Output: The output tensor.
|Constructors and Destructors|
| || |
|Public static functions|
Optional attribute setters for FusedResizeAndPadConv2D.
FusedResizeAndPadConv2D( const ::tensorflow::Scope & scope, ::tensorflow::Input input, ::tensorflow::Input size, ::tensorflow::Input paddings, ::tensorflow::Input filter, StringPiece mode, const gtl::ArraySlice< int > & strides, StringPiece padding )
FusedResizeAndPadConv2D( const ::tensorflow::Scope & scope, ::tensorflow::Input input, ::tensorflow::Input size, ::tensorflow::Input paddings, ::tensorflow::Input filter, StringPiece mode, const gtl::ArraySlice< int > & strides, StringPiece padding, const FusedResizeAndPadConv2D::Attrs & attrs )
::tensorflow::Node * node() const
Attrs ResizeAlignCorners( bool x )
© 2017 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.