pixel_shuffle_op.cc 5.8 KB
Newer Older
R
ruri 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
/*Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
    http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */

#include "paddle/fluid/operators/pixel_shuffle_op.h"
#include <memory>

namespace paddle {
namespace operators {

class PixelShuffleOp : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;

  void InferShape(framework::InferShapeContext* ctx) const override {
R
ruri 已提交
23 24 25 26 27 28
    PADDLE_ENFORCE_EQ(ctx->HasInput("X"), true,
                      platform::errors::NotFound(
                          "Input(X) of PixelShuffleOp should not be null."));
    PADDLE_ENFORCE_EQ(ctx->HasOutput("Out"), true,
                      platform::errors::NotFound(
                          "Output(Out) of PixelShuffleOp should not be null."));
R
ruri 已提交
29 30

    auto input_dims = ctx->GetInputDim("X");
R
ruri 已提交
31 32 33 34 35 36
    PADDLE_ENFORCE_EQ(
        input_dims.size(), 4,
        platform::errors::InvalidArgument(
            "Input should be a 4-D tensor of format [N, C, H, W], but got %u.",
            input_dims.size()));

R
ruri 已提交
37 38
    auto upscale_factor = ctx->Attrs().Get<int>("upscale_factor");

R
ruri 已提交
39 40 41 42 43
    PADDLE_ENFORCE_EQ(input_dims[1] % (upscale_factor * upscale_factor), 0,
                      platform::errors::InvalidArgument(
                          "The square of upscale_factor[%u] should divide the "
                          "number of channel[%u]",
                          input_dims[1], upscale_factor * upscale_factor));
R
ruri 已提交
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69

    auto output_dims = input_dims;
    output_dims[0] = input_dims[0];
    output_dims[1] = input_dims[1] / (upscale_factor * upscale_factor);
    output_dims[2] = input_dims[2] * upscale_factor;
    output_dims[3] = input_dims[3] * upscale_factor;
    ctx->SetOutputDim("Out", output_dims);
  }
};

class PixelShuffleOpMaker : public framework::OpProtoAndCheckerMaker {
 public:
  void Make() override {
    AddInput(
        "X",
        "(Tensor, default Tensor<float>), "
        "the input feature data of PixelShuffleOp, the layout is [N C H W].");
    AddOutput(
        "Out",
        "(Tensor, default Tensor<float>), the output of "
        "PixelShuffleOp. The layout is [N,C/factor^2,H*factor,W*factor].");
    AddAttr<int>("upscale_factor",
                 "the factor to increase spatial resolution by.")
        .SetDefault(1)
        .AddCustomChecker([](const int& upscale_factor) {
          PADDLE_ENFORCE_GE(upscale_factor, 1,
R
ruri 已提交
70 71
                            platform::errors::InvalidArgument(
                                "upscale_factor should be larger than 0."));
R
ruri 已提交
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
        });

    AddComment(R"DOC(
		Pixel Shuffle operator
		This operator rearranges elements in a tensor of shape :math:`(*, C \times r^2, H, W)`
    		to a tensor of shape :math:`(C, H \times r, W \times r)`.

		This is useful for implementing efficient sub-pixel convolution
    		with a stride of :math:`1/r`.

		Please refer to the paper:
		 `Real-Time Single Image and Video Super-Resolution Using an Efficient 
		 Sub-Pixel Convolutional Neural Network <https://arxiv.org/abs/1609.05158v2>`_
    		by Shi et. al (2016) for more details. 

        )DOC");
  }
};

H
hong 已提交
91 92
template <typename T>
class PixelShuffleGradMaker : public framework::SingleGradOpMaker<T> {
R
ruri 已提交
93
 public:
H
hong 已提交
94
  using framework::SingleGradOpMaker<T>::SingleGradOpMaker;
R
ruri 已提交
95

96
  void Apply(GradOpPtr<T> op) const override {
R
ruri 已提交
97
    op->SetType("pixel_shuffle_grad");
H
hong 已提交
98 99 100
    op->SetInput(framework::GradVarName("Out"), this->OutputGrad("Out"));
    op->SetAttrMap(this->Attrs());
    op->SetOutput(framework::GradVarName("X"), this->InputGrad("X"));
R
ruri 已提交
101 102 103 104 105 106 107 108
  }
};

class PixelShuffleGradOp : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;

  void InferShape(framework::InferShapeContext* ctx) const override {
R
ruri 已提交
109 110 111 112 113 114
    PADDLE_ENFORCE_EQ(
        ctx->HasInput(framework::GradVarName("Out")), true,
        platform::errors::NotFound("Input(Out@Grad) should not be null"));
    PADDLE_ENFORCE_EQ(
        ctx->HasOutput(framework::GradVarName("X")), true,
        platform::errors::NotFound("Output(X@Grad) should not be null"));
R
ruri 已提交
115 116

    auto do_dims = ctx->GetInputDim(framework::GradVarName("Out"));
R
ruri 已提交
117 118 119 120 121
    PADDLE_ENFORCE_EQ(
        do_dims.size(), 4,
        platform::errors::InvalidArgument(
            "Input should be a 4-D tensor of format [N, C, H, W], but got %u.",
            do_dims.size()));
R
ruri 已提交
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138

    auto upscale_factor = ctx->Attrs().Get<int>("upscale_factor");

    auto dx_dims = do_dims;
    dx_dims[0] = do_dims[0];
    dx_dims[1] = do_dims[1] * (upscale_factor * upscale_factor);
    dx_dims[2] = do_dims[2] / upscale_factor;
    dx_dims[3] = do_dims[3] / upscale_factor;
    ctx->SetOutputDim(framework::GradVarName("X"), dx_dims);
  }
};

}  // namespace operators
}  // namespace paddle

namespace ops = paddle::operators;
REGISTER_OPERATOR(pixel_shuffle, ops::PixelShuffleOp, ops::PixelShuffleOpMaker,
H
hong 已提交
139 140
                  ops::PixelShuffleGradMaker<paddle::framework::OpDesc>,
                  ops::PixelShuffleGradMaker<paddle::imperative::OpBase>);
R
ruri 已提交
141 142 143 144 145 146 147 148 149 150 151 152

REGISTER_OPERATOR(pixel_shuffle_grad, ops::PixelShuffleGradOp);

REGISTER_OP_CPU_KERNEL(
    pixel_shuffle,
    ops::PixelShuffleOpKernel<paddle::platform::CPUDeviceContext, float>,
    ops::PixelShuffleOpKernel<paddle::platform::CPUDeviceContext, double>);

REGISTER_OP_CPU_KERNEL(
    pixel_shuffle_grad,
    ops::PixelShuffleGradOpKernel<paddle::platform::CPUDeviceContext, float>,
    ops::PixelShuffleGradOpKernel<paddle::platform::CPUDeviceContext, double>);