conv2d_op.cc 5.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.

   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. */

H
hedaoyuan 已提交
15
#include "paddle/operators/gemm_conv2d_op.h"
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

namespace paddle {
namespace operators {

int outputSize(int input_size, int filter_size, int padding, int stride) {
  int output_size = (input_size - filter_size + 2 * padding) / stride + 1;
  return output_size;
}

class Conv2DOp : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;

 protected:
  void InferShape(const framework::InferShapeContext &ctx) const override {
31 32 33 34 35 36 37
    PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("Input"),
                            "Input(Input) of Conv2DOp should not be null.");
    PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("Filter"),
                            "Input(Filter) of Conv2DOp should not be null.");
    PADDLE_ENFORCE_NOT_NULL(ctx.OutputVar("Output"),
                            "Output(Output) of Conv2DOp should not be null.");

H
hedaoyuan 已提交
38 39
    auto in = ctx.Input<Tensor>("Input");
    auto filter = ctx.Input<Tensor>("Filter");
H
hedaoyuan 已提交
40
    auto out = ctx.Output<framework::LoDTensor>("Output");
H
hedaoyuan 已提交
41 42
    std::vector<int> strides = Attr<std::vector<int>>("strides");
    std::vector<int> paddings = Attr<std::vector<int>>("paddings");
H
hedaoyuan 已提交
43
    int groups = Attr<int>("groups");
H
hedaoyuan 已提交
44 45 46
    int input_channels = in->dims()[1];
    int output_channels = filter->dims()[0];

H
hedaoyuan 已提交
47
    PADDLE_ENFORCE_EQ(in->dims().size(), 4, "Conv2DOp input should be 4-D.");
48 49
    PADDLE_ENFORCE_EQ(filter->dims().size(), 4,
                      "Conv2DOp filter should be 4-D.");
H
hedaoyuan 已提交
50 51 52 53 54 55
    PADDLE_ENFORCE_EQ(input_channels, filter->dims()[1] * groups,
                      "The number of input channels should be equal to filter "
                      "channels * groups.");
    PADDLE_ENFORCE_EQ(
        output_channels % groups, 0,
        "The number of output channels should be divided by groups.");
56 57 58 59 60 61 62 63 64 65

    auto output_height =
        outputSize(in->dims()[2], filter->dims()[2], paddings[0], strides[0]);
    auto output_width =
        outputSize(in->dims()[3], filter->dims()[3], paddings[1], strides[1]);
    out->Resize(
        {in->dims()[0], filter->dims()[0], output_height, output_width});
  }
};

H
hedaoyuan 已提交
66
class Conv2DOpMaker : public framework::OpProtoAndCheckerMaker {
67
 public:
H
hedaoyuan 已提交
68
  Conv2DOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
69 70 71 72 73 74 75 76 77 78
      : OpProtoAndCheckerMaker(proto, op_checker) {
    AddInput(
        "Input",
        "The input tensor of convolution operator. "
        "The format of input tensor is NCHW. Where N is batch size, C is the "
        "number of channels, H and W is the height and width of image.");
    AddInput(
        "Filter",
        "The filter tensor of convolution operator."
        "The format of the filter tensor is MCHW, where M is the number of "
H
hedaoyuan 已提交
79 80 81 82
        "output image channels, C is the number of input image channels, "
        "H and W is height and width of filter. "
        "If the groups attribute is greater than 1, C equal the number of "
        "input image channels divided by the groups.");
83 84 85
    AddOutput("Output",
              "The output tensor of convolution operator."
              "The format of output tensor is also NCHW.");
H
hedaoyuan 已提交
86 87 88 89
    AddAttr<std::vector<int>>("strides", "strides of convolution operator.")
        .SetDefault({1, 1});
    AddAttr<std::vector<int>>("paddings", "paddings of convolution operator.")
        .SetDefault({0, 0});
H
hedaoyuan 已提交
90 91 92 93 94 95 96 97
    AddAttr<int>(
        "groups",
        "group size of convolution operator. "
        "Refer to grouped convolution in Alex Krizhevsky's paper: "
        "when group=2, the first half of the filters are only connected to the "
        "first half of the input channels, and the second half only connected "
        "to the second half.")
        .SetDefault(1);
H
hedaoyuan 已提交
98 99 100 101 102
    AddComment(R"DOC(
The convolution operation calculates the output based on the input, filter
and strides, paddings, groups parameters. The size of each dimension of the
parameters is checked in the infer-shape.
)DOC");
103 104 105 106 107 108 109 110
  }
};

class Conv2DOpGrad : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;

 protected:
H
hedaoyuan 已提交
111 112 113
  void InferShape(const framework::InferShapeContext &ctx) const override {
    auto in = ctx.Input<Tensor>("Input");
    auto filter = ctx.Input<Tensor>("Filter");
H
hedaoyuan 已提交
114 115 116 117
    auto d_in =
        ctx.Output<framework::LoDTensor>(framework::GradVarName("Input"));
    auto d_filter =
        ctx.Output<framework::LoDTensor>(framework::GradVarName("Filter"));
118 119
    if (d_in) d_in->Resize(in->dims());
    if (d_filter) d_filter->Resize(filter->dims());
H
hedaoyuan 已提交
120
  }
121 122 123 124 125 126
};

}  // namespace operators
}  // namespace paddle

namespace ops = paddle::operators;
H
hedaoyuan 已提交
127
REGISTER_OP(conv2d, ops::Conv2DOp, ops::Conv2DOpMaker, conv2d_grad,
128 129 130
            ops::Conv2DOpGrad);

REGISTER_OP_CPU_KERNEL(
H
hedaoyuan 已提交
131
    conv2d, ops::GemmConv2DKernel<paddle::platform::CPUPlace, float>);
H
hedaoyuan 已提交
132
REGISTER_OP_CPU_KERNEL(
H
hedaoyuan 已提交
133
    conv2d_grad, ops::GemmConvGrad2DKernel<paddle::platform::CPUPlace, float>);