GemmConvOp.cpp 12.2 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. */

15
#include "ConvOp.h"
16
#include "GemmFunctor.h"
17
#include "Im2Col.h"
18 19 20 21 22
#include "paddle/math/MemoryHandle.h"

namespace paddle {

/*
23
 * \brief Forward calculation of convolution.
24 25 26 27 28 29 30 31
 */
template <DeviceType Device>
class GemmConvFunction : public ConvFunctionBase {
public:
  void init(const FuncConfig& config) override {
    ConvFunctionBase::init(config);
  }

L
liaogang 已提交
32
  void check(const BufferArgs& inputs, const BufferArgs& outputs) override {
H
hedaoyuan 已提交
33 34 35 36 37 38
    const TensorShape& input = inputs[0].shape();
    const TensorShape& filter = inputs[1].shape();
    const TensorShape& output = outputs[0].shape();
    checkShape(input, filter, output);
  }

39
  void calc(const BufferArgs& inputs, const BufferArgs& outputs) override {
40 41
    CHECK_EQ(numInputs_, inputs.size());
    CHECK_EQ(numOutputs_, outputs.size());
H
hedaoyuan 已提交
42
    check(inputs, outputs);
43 44 45 46 47 48 49 50 51 52 53 54
    // TODO(hedaoyuan): Need to define some index macros,
    // to avoid useing 0 and 1.
    const TensorShape& input = inputs[0].shape();
    const TensorShape& filter = inputs[1].shape();
    const TensorShape& output = outputs[0].shape();

    real beta;
    if (outputs[0].getArgType() == ADD_TO) {
      beta = 1.0;
    } else {
      beta = 0.0;
    }
55

H
hedaoyuan 已提交
56 57 58 59 60 61 62 63 64
    size_t batchSize = input[0];
    size_t inputChannels = input[1];
    size_t inputHeight = input[2];
    size_t inputWidth = input[3];
    size_t filterHeight = getFilterHeight(filter);
    size_t filterWidth = getFilterWidth(filter);
    size_t outputChannels = output[1];
    size_t outputHeight = output[2];
    size_t outputWidth = output[3];
65 66 67 68

    real* inputData = inputs[0].data<real>();
    real* filterData = inputs[1].data<real>();
    real* outputData = outputs[0].data<real>();
69
    bool needIm2col = isNeedIm2col(filter);
70

71 72 73
    TensorShape imShape =
        TensorShape({inputChannels / groups_, inputHeight, inputWidth});

74
    TensorShape colShape;
75
    real* colData = NULL;
76

77
    if (needIm2col) {
78 79 80 81 82 83 84 85
      colShape = TensorShape({inputChannels / groups_,
                              filterHeight,
                              filterWidth,
                              outputHeight,
                              outputWidth});
      resizeBuffer<Device>(colShape.getElements());
      colData = reinterpret_cast<real*>(memory_->getBuf());
    }
86

87 88
    Im2ColFunctor<kCFO, Device, real> im2col;
    size_t inputOffset = imShape.getElements();
89 90
    size_t outputOffset =
        (outputChannels / groups_) * outputHeight * outputWidth;
H
hedaoyuan 已提交
91 92
    size_t filterOffset = filter.getElements() / groups_;

93
    for (size_t i = 0; i < batchSize; i++) {
94
      for (size_t g = 0; g < groups_; g++) {
95
        if (needIm2col) {
96 97 98 99 100 101 102
          im2col(inputData + g * inputOffset,
                 imShape,
                 colData,
                 colShape,
                 strideH(),
                 strideW(),
                 paddingH(),
X
xzl 已提交
103 104 105
                 paddingW(),
                 dilationH(),
                 dilationW());
106 107
        } else {
          colData = inputData + g * inputOffset;
108
        }
H
Bug fix  
hedaoyuan 已提交
109
        int M = outputChannels / groups_;
110
        int N = outputHeight * outputWidth;
H
Bug fix  
hedaoyuan 已提交
111
        int K = inputChannels / groups_ * filterHeight * filterWidth;
H
hedaoyuan 已提交
112 113 114 115 116 117 118 119 120 121 122 123 124
        BlasGemm<Device, real>::compute(false,
                                        false,
                                        M,
                                        N,
                                        K,
                                        1.0f,
                                        filterData + g * filterOffset,
                                        K,
                                        colData,
                                        N,
                                        beta,
                                        outputData + g * outputOffset,
                                        N);
125
      }
H
hedaoyuan 已提交
126 127
      inputData += inputChannels * inputHeight * inputWidth;
      outputData += outputChannels * outputHeight * outputWidth;
128 129 130 131
    }
  }
};

132 133 134 135 136 137 138 139 140 141
/*
 * \brief Backward input calculation of convolution.
 */
template <DeviceType Device>
class GemmConvGradInputFunction : public ConvFunctionBase {
public:
  void init(const FuncConfig& config) override {
    ConvFunctionBase::init(config);
  }

L
liaogang 已提交
142
  void check(const BufferArgs& inputs, const BufferArgs& outputs) override {
H
hedaoyuan 已提交
143 144 145 146 147 148
    const TensorShape& output = inputs[0].shape();
    const TensorShape& filter = inputs[1].shape();
    const TensorShape& input = outputs[0].shape();
    checkShape(input, filter, output);
  }

149 150 151
  void calc(const BufferArgs& inputs, const BufferArgs& outputs) override {
    CHECK_EQ(numInputs_, inputs.size());
    CHECK_EQ(numOutputs_, outputs.size());
H
hedaoyuan 已提交
152
    check(inputs, outputs);
H
hedaoyuan 已提交
153 154 155
    // Since the implementation of Col2ImFunctor is ADD_TO,
    // this function only supports ADD_TO mode.
    CHECK_EQ(outputs[0].getArgType(), ADD_TO);
156
    const TensorShape& output = inputs[0].shape();
157
    const TensorShape& filter = inputs[1].shape();
158 159 160 161 162 163
    const TensorShape& input = outputs[0].shape();

    size_t batchSize = input[0];
    size_t inputChannels = input[1];
    size_t inputHeight = input[2];
    size_t inputWidth = input[3];
H
hedaoyuan 已提交
164 165
    size_t filterHeight = getFilterHeight(filter);
    size_t filterWidth = getFilterWidth(filter);
166 167 168 169 170 171 172
    size_t outputChannels = output[1];
    size_t outputHeight = output[2];
    size_t outputWidth = output[3];

    real* outputGrad = inputs[0].data<real>();
    real* filterData = inputs[1].data<real>();
    real* inputGrad = outputs[0].data<real>();
173
    bool needIm2col = isNeedIm2col(filter);
174

175 176 177
    TensorShape imShape =
        TensorShape({inputChannels / groups_, inputHeight, inputWidth});

178
    TensorShape colShape;
179
    real* colData = NULL;
180

181
    if (needIm2col) {
182 183 184 185 186 187 188 189
      colShape = TensorShape({inputChannels / groups_,
                              filterHeight,
                              filterWidth,
                              outputHeight,
                              outputWidth});
      resizeBuffer<Device>(colShape.getElements());
      colData = reinterpret_cast<real*>(memory_->getBuf());
    }
190

191 192
    Col2ImFunctor<kCFO, Device, real> col2im;
    size_t inputOffset = imShape.getElements();
H
format  
hedaoyuan 已提交
193
    size_t outputOffset =
194 195 196 197 198 199 200 201
        (outputChannels / groups_) * outputHeight * outputWidth;
    size_t filterOffset = filter.getElements() / groups_;

    for (size_t i = 0; i < batchSize; i++) {
      for (size_t g = 0; g < groups_; g++) {
        int K = outputChannels / groups_;
        int N = outputHeight * outputWidth;
        int M = inputChannels / groups_ * filterHeight * filterWidth;
202
        real scale = 0.0f;
203 204
        if (!needIm2col) {
          colData = inputGrad + g * inputOffset;
205 206
          scale = 1.0f;
        }
H
hedaoyuan 已提交
207 208 209 210 211 212 213 214 215 216 217 218 219
        BlasGemm<Device, real>::compute(true,
                                        false,
                                        M,
                                        N,
                                        K,
                                        1.0f,
                                        filterData + g * filterOffset,
                                        M,
                                        outputGrad + g * outputOffset,
                                        N,
                                        scale,
                                        colData,
                                        N);
220
        if (needIm2col) {
221 222
          col2im(inputGrad + g * inputOffset,
                 imShape,
223
                 colData,
224 225 226 227
                 colShape,
                 strideH(),
                 strideW(),
                 paddingH(),
X
xzl 已提交
228 229 230
                 paddingW(),
                 dilationH(),
                 dilationW());
231
        }
232 233 234 235
      }
      inputGrad += inputChannels * inputHeight * inputWidth;
      outputGrad += outputChannels * outputHeight * outputWidth;
    }
236 237 238 239 240 241 242 243 244 245 246 247 248
  }
};

/*
 * \brief Backward filter calculation of convolution.
 */
template <DeviceType Device>
class GemmConvGradFilterFunction : public ConvFunctionBase {
public:
  void init(const FuncConfig& config) override {
    ConvFunctionBase::init(config);
  }

L
liaogang 已提交
249
  void check(const BufferArgs& inputs, const BufferArgs& outputs) override {
H
hedaoyuan 已提交
250 251 252 253 254 255
    const TensorShape& output = inputs[0].shape();
    const TensorShape& input = inputs[1].shape();
    const TensorShape& filter = outputs[0].shape();
    checkShape(input, filter, output);
  }

256 257 258
  void calc(const BufferArgs& inputs, const BufferArgs& outputs) override {
    CHECK_EQ(numInputs_, inputs.size());
    CHECK_EQ(numOutputs_, outputs.size());
H
hedaoyuan 已提交
259
    check(inputs, outputs);
260
    const TensorShape& output = inputs[0].shape();
261
    const TensorShape& input = inputs[1].shape();
262 263
    const TensorShape& filter = outputs[0].shape();

264 265 266 267 268 269 270
    real beta;
    if (outputs[0].getArgType() == ADD_TO) {
      beta = 1.0;
    } else {
      beta = 0.0;
    }

271 272 273 274
    size_t batchSize = input[0];
    size_t inputChannels = input[1];
    size_t inputHeight = input[2];
    size_t inputWidth = input[3];
H
hedaoyuan 已提交
275 276
    size_t filterHeight = getFilterHeight(filter);
    size_t filterWidth = getFilterWidth(filter);
277 278 279 280 281 282 283
    size_t outputChannels = output[1];
    size_t outputHeight = output[2];
    size_t outputWidth = output[3];

    real* outputGrad = inputs[0].data<real>();
    real* inputData = inputs[1].data<real>();
    real* filterGrad = outputs[0].data<real>();
284
    bool needIm2col = isNeedIm2col(filter);
285

286 287 288
    TensorShape imShape =
        TensorShape({inputChannels / groups_, inputHeight, inputWidth});

289
    TensorShape colShape;
290
    real* colData = NULL;
291

292
    if (needIm2col) {
293 294 295 296 297 298 299 300
      colShape = TensorShape({inputChannels / groups_,
                              filterHeight,
                              filterWidth,
                              outputHeight,
                              outputWidth});
      resizeBuffer<Device>(colShape.getElements());
      colData = reinterpret_cast<real*>(memory_->getBuf());
    }
301

302 303
    Im2ColFunctor<kCFO, Device, real> im2col;
    size_t inputOffset = imShape.getElements();
304 305 306 307 308
    size_t outputOffset =
        (outputChannels / groups_) * outputHeight * outputWidth;
    size_t filterOffset = filter.getElements() / groups_;
    for (size_t i = 0; i < batchSize; i++) {
      for (size_t g = 0; g < groups_; g++) {
309
        if (needIm2col) {
310 311 312 313 314 315 316
          im2col(inputData + g * inputOffset,
                 imShape,
                 colData,
                 colShape,
                 strideH(),
                 strideW(),
                 paddingH(),
X
xzl 已提交
317 318 319
                 paddingW(),
                 dilationH(),
                 dilationW());
320 321
        } else {
          colData = inputData + g * inputOffset;
322
        }
323 324 325
        int M = outputChannels / groups_;
        int K = outputHeight * outputWidth;
        int N = inputChannels / groups_ * filterHeight * filterWidth;
H
hedaoyuan 已提交
326 327 328 329 330 331 332 333 334 335 336 337 338
        BlasGemm<Device, real>::compute(false,
                                        true,
                                        M,
                                        N,
                                        K,
                                        1.0f,
                                        outputGrad + g * outputOffset,
                                        K,
                                        colData,
                                        K,
                                        i == 0 ? beta : 1.0f,
                                        filterGrad + g * filterOffset,
                                        N);
339
      }
340 341
      inputData += inputChannels * inputHeight * inputWidth;
      outputGrad += outputChannels * outputHeight * outputWidth;
342
    }
343 344 345
  }
};

346
REGISTER_TYPED_FUNC(GemmConv, CPU, GemmConvFunction);
347 348
REGISTER_TYPED_FUNC(GemmConvGradInput, CPU, GemmConvGradInputFunction);
REGISTER_TYPED_FUNC(GemmConvGradFilter, CPU, GemmConvGradFilterFunction);
349
#ifdef PADDLE_WITH_CUDA
350
REGISTER_TYPED_FUNC(GemmConv, GPU, GemmConvFunction);
351 352
REGISTER_TYPED_FUNC(GemmConvGradInput, GPU, GemmConvGradInputFunction);
REGISTER_TYPED_FUNC(GemmConvGradFilter, GPU, GemmConvGradFilterFunction);
H
hedaoyuan 已提交
353
#endif
354 355

}  // namespace paddle