pad_op.cc 4.5 KB
Newer Older
W
wanghaoshuang 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 23 24 25 26
/* 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. */

#include "paddle/operators/pad_op.h"

namespace paddle {
namespace operators {

using framework::Tensor;

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

 protected:
Q
Qiao Longfei 已提交
27 28 29 30 31 32 33
  void InferShape(framework::InferShapeContextBase* ctx) const override {
    PADDLE_ENFORCE(ctx->HasInput("X"), "Input(X) of PadOp should not be null.");
    PADDLE_ENFORCE(ctx->HasOutput("Out"),
                   "Output(Out) of PadOp should not be null.");

    auto x_dim = ctx->GetInputDim("X");
    auto paddings = ctx->Attrs().Get<std::vector<int>>("paddings");
W
wanghaoshuang 已提交
34
    PADDLE_ENFORCE_EQ(x_dim.size() * 2, int64_t(paddings.size()),
W
wanghaoshuang 已提交
35 36
                      "Size of paddings should be equal to 2 * dimension size "
                      "of input tensor.");
W
wanghaoshuang 已提交
37
    std::vector<int64_t> out_dims(x_dim.size());
W
wanghaoshuang 已提交
38 39
    for (int i = 0; i < x_dim.size(); ++i) {
      out_dims[i] = x_dim[i] + paddings[i * 2] + paddings[i * 2 + 1];
W
wanghaoshuang 已提交
40
    }
Q
Qiao Longfei 已提交
41
    ctx->SetOutputDim("Out", framework::make_ddim(out_dims));
D
Fix bug  
dangqingqing 已提交
42 43 44
    if (out_dims[0] == x_dim[0]) {
      // Only pass LoD when the first dimension is equal between
      // output and input.
Q
Qiao Longfei 已提交
45
      ctx->ShareLoD("X", /*->*/ "Out");
D
Fix bug  
dangqingqing 已提交
46
    }
W
wanghaoshuang 已提交
47 48 49
  }
};

W
wanghaoshuang 已提交
50
class PadOpMaker : public framework::OpProtoAndCheckerMaker {
W
wanghaoshuang 已提交
51
 public:
Q
Qiao Longfei 已提交
52
  PadOpMaker(framework::OpProto* proto, framework::OpAttrChecker* op_checker)
W
wanghaoshuang 已提交
53
      : OpProtoAndCheckerMaker(proto, op_checker) {
W
wanghaoshuang 已提交
54 55 56 57 58
    AddInput("X",
             "The input of pad op. "
             "The input should be a k-D tensor(k > 0 and k < 7)");
    AddOutput("Out",
              "The output of pad op."
59
              "A tensor with the same shape as X.");
W
wanghaoshuang 已提交
60
    AddComment(R"DOC(
W
wanghaoshuang 已提交
61 62 63 64 65 66 67
Pad input into output, as specified by paddings and pad_value. The input should be a k-D tensor(k > 0 and k < 7). As an example:

Given:

X = [[1, 2],
   [3, 4]]

Q
Qiao Longfei 已提交
68
and
W
wanghaoshuang 已提交
69

W
wanghaoshuang 已提交
70
paddings = [0, 1, 1, 2]
W
wanghaoshuang 已提交
71 72 73

and

Q
Qiao Longfei 已提交
74 75 76
pad_value = 0

then we get
W
wanghaoshuang 已提交
77 78 79 80

Out = [[0, 1, 2, 0, 0]
       [0, 3, 4, 0, 0]
       [0, 0, 0, 0, 0]]
W
wanghaoshuang 已提交
81
)DOC");
W
wanghaoshuang 已提交
82 83
    AddAttr<std::vector<int>>(
        "paddings",
W
wanghaoshuang 已提交
84 85
        "A list<int> to describes padding rules for each dimension."
        " For 2-D image tensor, paddings=[0, 1, 2, 3] means"
W
wanghaoshuang 已提交
86
        " padding 0 row to top, 1 row to bottom, 2 columns to left"
W
wanghaoshuang 已提交
87 88
        " and 3 columns to right.Size of paddings should be equal to"
        " 2 * dimension size of input tensor.");
W
wanghaoshuang 已提交
89 90
    AddAttr<float>("pad_value",
                   "(float) default to 0; "
W
wanghaoshuang 已提交
91
                   "The value to fill padded areas.")
W
wanghaoshuang 已提交
92 93 94 95 96 97 98 99 100
        .SetDefault(0.0f);
  }
};

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

 protected:
Q
Qiao Longfei 已提交
101 102 103 104 105 106 107 108
  void InferShape(framework::InferShapeContextBase* ctx) const override {
    PADDLE_ENFORCE(ctx->HasInput("X"), "Input(X) should not be null");
    PADDLE_ENFORCE(ctx->HasInput(framework::GradVarName("Out")),
                   "Input(Out@GRAD) should not be null");
    auto x_dims = ctx->GetInputDim("X");
    auto x_grad_name = framework::GradVarName("X");
    if (ctx->HasOutput(x_grad_name)) {
      ctx->SetOutputDim(x_grad_name, x_dims);
W
wanghaoshuang 已提交
109
    }
W
wanghaoshuang 已提交
110 111 112
  }
};

113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
class PadOpGradMaker : public framework::SingleGradOpDescMaker {
 protected:
  framework::OpDescBind Apply() const override {
    framework::OpDescBind bind;
    bind.SetInput("X", Input("X"));
    bind.SetInput(framework::GradVarName("Out"), OutputGrad("Out"));
    bind.SetOutput(framework::GradVarName("X"), InputGrad("X"));
    bind.SetAttrMap(Attrs());
    return bind;
  }

 public:
  using framework::SingleGradOpDescMaker::SingleGradOpDescMaker;
};

W
wanghaoshuang 已提交
128 129 130 131
}  // namespace operators
}  // namespace paddle

namespace ops = paddle::operators;
132 133 134

REGISTER_OPERATOR(pad, ops::PadOp, ops::PadOpMaker, ops::PadOpGradMaker);
REGISTER_OPERATOR(pad_grad, ops::PadOpGrad);
W
wanghaoshuang 已提交
135 136 137
REGISTER_OP_CPU_KERNEL(pad, ops::PadKernel<paddle::platform::CPUPlace, float>);
REGISTER_OP_CPU_KERNEL(pad_grad,
                       ops::PadGradKernel<paddle::platform::CPUPlace, float>);