scale_op.cc 4.5 KB
Newer Older
1
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
Y
Yu Yang 已提交
2

L
Luo Tao 已提交
3 4 5
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
Y
Yu Yang 已提交
6

L
Luo Tao 已提交
7
    http://www.apache.org/licenses/LICENSE-2.0
Y
Yu Yang 已提交
8

L
Luo Tao 已提交
9 10 11 12 13
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. */
Y
Yu Yang 已提交
14

15
#include <string>
16

17 18
#include "paddle/fluid/framework/infershape_utils.h"
#include "paddle/fluid/framework/op_registry.h"
19
#include "paddle/fluid/platform/float16.h"
20 21
#include "paddle/phi/core/infermeta_utils.h"
#include "paddle/phi/infermeta/unary.h"
W
wanghuancoder 已提交
22

Y
Yu Yang 已提交
23 24 25 26 27
namespace paddle {
namespace operators {

class ScaleOp : public framework::OperatorWithKernel {
 public:
28
  using framework::OperatorWithKernel::OperatorWithKernel;
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

  framework::OpKernelType GetExpectedKernelType(
      const framework::ExecutionContext &ctx) const override {
    auto input_data_type =
        framework::OperatorWithKernel::IndicateVarDataType(ctx, "X");

#ifdef PADDLE_WITH_MKLDNN
    if (this->CanMKLDNNBeUsed(ctx, input_data_type)) {
      return framework::OpKernelType(input_data_type, ctx.GetPlace(),
                                     framework::DataLayout::kMKLDNN,
                                     framework::LibraryType::kMKLDNN);
    }
#endif
    return framework::OpKernelType(input_data_type, ctx.GetPlace());
  }
Y
Yu Yang 已提交
44 45 46 47
};

class ScaleOpMaker : public framework::OpProtoAndCheckerMaker {
 public:
Y
Yu Yang 已提交
48
  void Make() override {
49
    AddInput("X", "(Tensor) Input tensor of scale operator.");
50 51 52 53 54
    AddInput("ScaleTensor",
             "(Tensor) If provided, use this as "
             "scale factor, this has a higher priority than "
             "attr(scale), the shape of this tensor MUST BE 1.")
        .AsDispensable();
55 56
    AddOutput("Out", "(Tensor) Output tensor of scale operator.");
    AddComment(R"DOC(
Y
yi.wu 已提交
57 58
**Scale operator**

S
sneaxiy 已提交
59
Apply scaling and bias addition to the input tensor.
Y
Yu Yang 已提交
60

S
sneaxiy 已提交
61 62 63 64 65 66 67
if bias_after_scale=True:

$$Out = scale*X + bias$$

else:

$$Out = scale*(X + bias)$$
Y
Yu Yang 已提交
68
)DOC");
Y
yi.wu 已提交
69
    AddAttr<float>("scale", "The scaling factor of the scale operator.")
C
caoying03 已提交
70
        .SetDefault(1.0);
S
sneaxiy 已提交
71
    AddAttr<float>("bias", "The bias of the scale operator.").SetDefault(0.0);
S
sneaxiy 已提交
72 73 74 75 76
    AddAttr<bool>(
        "bias_after_scale",
        "Apply bias addition after or before scaling. It is useful for "
        "numeric stability in some circumstances.")
        .SetDefault(true);
77 78
    AddAttr<bool>("use_mkldnn",
                  "(bool, default false) Only used in mkldnn kernel")
79 80
        .SetDefault(false)
        .AsExtra();
Y
Yu Yang 已提交
81 82 83
  }
};

84 85
class ScaleOpVarTypeInference : public framework::VarTypeInference {
 public:
M
minqiyang 已提交
86
  void operator()(framework::InferVarTypeContext *ctx) const override {
87
    ctx->SyncTypeAndDataType("X", "Out");
88 89 90
  }
};

H
hong 已提交
91 92
template <typename T>
class ScaleGradMaker : public framework::SingleGradOpMaker<T> {
Y
Yu Yang 已提交
93
 public:
H
hong 已提交
94
  using framework::SingleGradOpMaker<T>::SingleGradOpMaker;
95

96
  void Apply(GradOpPtr<T> grad_op) const override {
Y
Yu Yang 已提交
97
    grad_op->SetType("scale");
H
hong 已提交
98
    grad_op->SetInput("X", this->OutputGrad("Out"));
99 100 101
    if (this->HasInput("ScaleTensor") > 0) {
      grad_op->SetInput("ScaleTensor", this->Input("ScaleTensor"));
    }
H
hong 已提交
102
    grad_op->SetOutput("Out", this->InputGrad("X"));
J
Jiabin Yang 已提交
103
    VLOG(6) << "Finish SetOutput";
H
hong 已提交
104
    grad_op->SetAttr("scale", this->GetAttr("scale"));
J
Jiabin Yang 已提交
105
    VLOG(6) << "Finish Set Attr scale";
S
sneaxiy 已提交
106
    grad_op->SetAttr("bias", 0.0f);
J
Jiabin Yang 已提交
107
    VLOG(6) << "Finish Set Attr bias";
S
sneaxiy 已提交
108
    grad_op->SetAttr("bias_after_scale", true);
J
Jiabin Yang 已提交
109 110 111
    VLOG(6) << "Finish Set Attr bias_after_scale";
    if (grad_op->HasAttr("use_mkldnn")) {
      VLOG(6) << "Finish Check Attr use_mkldnn";
112
      grad_op->SetAttr("use_mkldnn", this->GetAttr("use_mkldnn"));
J
Jiabin Yang 已提交
113 114 115
      VLOG(6) << "Finish Set Attr use_mkldnn";
    }
    VLOG(6) << "Finish Apply";
Y
Yu Yang 已提交
116 117 118
  }
};

119
DECLARE_INPLACE_OP_INFERER(ScaleOpInplaceInferer, {"X", "Out"});
Y
Yu Yang 已提交
120 121 122 123 124
}  // namespace operators
}  // namespace paddle

namespace ops = paddle::operators;

125 126
DECLARE_INFER_SHAPE_FUNCTOR(scale, ScaleInferShapeFunctor,
                            PD_INFER_META(phi::UnchangedInferMeta));
H
hong 已提交
127 128 129
REGISTER_OPERATOR(scale, ops::ScaleOp, ops::ScaleOpMaker,
                  ops::ScaleGradMaker<paddle::framework::OpDesc>,
                  ops::ScaleGradMaker<paddle::imperative::OpBase>,
130 131
                  ScaleInferShapeFunctor, ops::ScaleOpVarTypeInference,
                  ops::ScaleOpInplaceInferer);