1 | /* Copyright 2015 The TensorFlow Authors. All Rights Reserved. |
2 | |
3 | Licensed under the Apache License, Version 2.0 (the "License"); |
4 | you may not use this file except in compliance with the License. |
5 | You may obtain a copy of the License at |
6 | |
7 | http://www.apache.org/licenses/LICENSE-2.0 |
8 | |
9 | Unless required by applicable law or agreed to in writing, software |
10 | distributed under the License is distributed on an "AS IS" BASIS, |
11 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
12 | See the License for the specific language governing permissions and |
13 | limitations under the License. |
14 | ==============================================================================*/ |
15 | |
16 | #ifndef TENSORFLOW_CORE_KERNELS_AVGPOOLING_OP_H_ |
17 | #define TENSORFLOW_CORE_KERNELS_AVGPOOLING_OP_H_ |
18 | // Functor definition for AvgPoolingOp, must be compilable by nvcc. |
19 | |
20 | #include "tensorflow/core/framework/tensor_types.h" |
21 | #include "tensorflow/core/kernels/eigen_pooling.h" |
22 | #include "tensorflow/core/platform/types.h" |
23 | |
24 | namespace tensorflow { |
25 | namespace functor { |
26 | |
27 | template <typename Device, typename T> |
28 | struct SpatialAvgPooling { |
29 | void operator()(const Device& d, typename TTypes<T, 4>::Tensor output, |
30 | typename TTypes<T, 4>::ConstTensor input, int window_rows, |
31 | int window_cols, int row_stride, int col_stride, |
32 | const Eigen::PaddingType& padding) { |
33 | MaybeWith32BitIndexing<Device>( |
34 | [&](auto output32, auto input32) { |
35 | // Because we swap the layout, we swap the row/cols as well. |
36 | output32.swap_layout().device(d) = Eigen::SpatialAvgPooling( |
37 | input32.swap_layout(), window_cols, window_rows, col_stride, |
38 | row_stride, padding); |
39 | }, |
40 | output, input); |
41 | } |
42 | }; |
43 | |
44 | } // namespace functor |
45 | |
46 | typedef Eigen::GpuDevice GPUDevice; |
47 | |
48 | // Launch a custom GPU kernels from Yanqing for the avgpooling backward |
49 | // operation that works NHWC data formats. Arguments: |
50 | // top_diff: backprop to the output of the pooling layer |
51 | // num: number of input batches |
52 | // height: input height |
53 | // width: input width |
54 | // channels: number of input channels |
55 | // pooled_height: the height of the output to the pooling layer |
56 | // pooled_width: the width of the output to the pooling layer |
57 | // kernel_h: the height of the pooling kernel |
58 | // kernel_w: the width of the pooling kernel |
59 | // stride_h: the height of the vertical stride |
60 | // stride_w: the width of the horizontal stride |
61 | // pad_t: padding size to the top side |
62 | // pad_l: padding size to the left side |
63 | // bottom_diff: backprop to the input of the pooling layer. |
64 | template <typename T> |
65 | bool RunAvePoolBackwardNHWC(const T* const top_diff, const int num, |
66 | const int height, const int width, |
67 | const int channels, const int pooled_height, |
68 | const int pooled_width, const int kernel_h, |
69 | const int kernel_w, const int stride_h, |
70 | const int stride_w, const int pad_t, |
71 | const int pad_l, T* const bottom_diff, |
72 | const GPUDevice& d); |
73 | |
74 | } // namespace tensorflow |
75 | |
76 | #endif // TENSORFLOW_CORE_KERNELS_AVGPOOLING_OP_H_ |
77 | |