reduce_op.h 27.4 KB
Newer Older
1
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
G
guosheng 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
G
guosheng 已提交
6

L
Luo Tao 已提交
7
    http://www.apache.org/licenses/LICENSE-2.0
G
guosheng 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. */
G
guosheng 已提交
14 15 16

#pragma once

17
#include <algorithm>
18
#include <set>
19
#include <string>
W
whs 已提交
20
#include <vector>
21
#include "paddle/fluid/framework/data_type_transform.h"
22
#include "paddle/fluid/framework/tensor_util.h"
23
#include "paddle/fluid/operators/cast_op.h"
24
#include "paddle/fluid/operators/math/math_function.h"
W
Wu Yi 已提交
25
#include "paddle/fluid/operators/reduce_ops/reduce_op_function.h"
26 27 28
#if defined(__HIPCC__) || defined(__NVCC__)
#include "paddle/fluid/operators/reduce_ops/reduce_op.cu.h"
#endif
G
guosheng 已提交
29 30 31 32

namespace paddle {
namespace operators {

33 34
#define HANDLE_DIM(NDIM, RDIM)                                            \
  if (ndim == NDIM && rdim == RDIM) {                                     \
35
    ReduceFunctor<DeviceContext, OutT, NDIM, RDIM, Functor>(              \
36 37
        context.template device_context<DeviceContext>(), *input, output, \
        dims, keep_dim);                                                  \
W
whs 已提交
38 39
  }

40
using Tensor = framework::Tensor;
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
using DDim = framework::DDim;

inline void GetShuffledDim(const DDim& src_dims, DDim* dst_dims,
                           const std::vector<int>& reduced_dims,
                           std::vector<int>* perm_axis) {
  // check if it's a reduced dim
  std::vector<bool> src_dims_check(src_dims.size(), false);
  size_t src_size = src_dims.size();
  size_t reduce_size = reduced_dims.size();
  for (size_t i = 0; i < reduce_size; ++i) {
    dst_dims->at(src_size - reduce_size + i) = src_dims[reduced_dims[i]];
    (*perm_axis)[src_size - reduce_size + i] = reduced_dims[i];
    src_dims_check[reduced_dims[i]] = true;
  }

  size_t offset = 0;
  for (size_t i = 0; i < src_dims_check.size(); ++i) {
    bool is_reduced = src_dims_check[i];
    if (!is_reduced) {
      (*perm_axis)[offset] = i;
      dst_dims->at(offset++) = src_dims[i];
    }
  }
}

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
static inline std::vector<int> GetReduceDim(const std::vector<int>& dims,
                                            int dim_size, bool reduce_all) {
  std::vector<int> reduce_dims;
  if (reduce_all) {
    reduce_dims.resize(dim_size);
    int reduce_size = reduce_dims.size();
    for (int i = 0; i < reduce_size; ++i) {
      reduce_dims[i] = i;
    }
  } else {
    for (auto e : dims) {
      PADDLE_ENFORCE_LT(e, dim_size,
                        paddle::platform::errors::InvalidArgument(
                            "ReduceOp: invalid axis, when x_dims is %d, "
                            "axis[i] should less than x_dims, but got %d.",
                            dim_size, e));
      reduce_dims.push_back(e >= 0 ? e : e + dim_size);
    }
  }
  return reduce_dims;
}
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
template <typename DeviceContext, typename OutT>
void GetShuffledInput(const framework::ExecutionContext& context,
                      const Tensor* input, Tensor* shuffled_input,
                      const std::vector<int>& dims) {
  DDim shuffled_dims(input->dims());
  std::vector<int> perm_axis(input->dims().size());
  GetShuffledDim(input->dims(), &shuffled_dims, dims, &perm_axis);

  shuffled_input->Resize(shuffled_dims);
  shuffled_input->mutable_data<OutT>(context.GetPlace());

  math::TransposeNormal<DeviceContext, OutT> trans;
  trans(context.template device_context<DeviceContext>(), *input,
        shuffled_input, perm_axis);
}

inline void GetOriginDimFromShuffled(const DDim& src_dim,
                                     const std::vector<int>& dims,
                                     std::vector<int>* origin_dim) {
  DDim shuffled_dims(src_dim);
  size_t n = src_dim.size();
  std::vector<int> perm_axis(n);
  GetShuffledDim(src_dim, &shuffled_dims, dims, &perm_axis);
  for (size_t i = 0; i < n; ++i) {
    (*origin_dim)[perm_axis[i]] = i;
  }
}

template <typename DeviceContext, typename OutT, typename Functor>
void HandleLargeDim(const framework::ExecutionContext& context,
                    const Tensor* input, Tensor* output,
                    const std::vector<int>& dims, bool keep_dim) {
  //  shuffle the reduced dim to the end
  Tensor shuffled_input;
  GetShuffledInput<DeviceContext, OutT>(context, input, &shuffled_input, dims);

  // transpose to 2D tensor whose shape is {unreduced, reduced}.
  const int64_t unreduced = output->numel();
  const int64_t reduced = shuffled_input.numel() / unreduced;
  shuffled_input.Resize({unreduced, reduced});
  DDim output_dim = output->dims();
  output->Resize({unreduced});
  ReduceFunctor<DeviceContext, OutT, 2, 1, Functor>(
      context.template device_context<DeviceContext>(), shuffled_input, output,
      {1}, keep_dim);
  output->Resize(output_dim);
}

template <typename DeviceContext, typename T, typename Functor>
void HandleLargeDimGrad(const framework::ExecutionContext& context,
                        const framework::Tensor* x,
                        const framework::Tensor* out,
                        const framework::Tensor* dout, framework::Tensor* dx,
                        const std::vector<int>& dims) {
  const int64_t unreduced = out->numel();
  const int64_t reduced = x->numel() / unreduced;
  DDim out_dim(out->dims());
  DDim x_dim(x->dims());
  // transpose and reshape X
  Tensor shuffled_x;
  GetShuffledInput<DeviceContext, T>(context, x, &shuffled_x, dims);
  DDim shuffled_dim = shuffled_x.dims();
  shuffled_x.Resize({unreduced, reduced});
  // reshape dX {unreduced, reduced}
  dx->Resize({unreduced, reduced});
  ReduceGradFunctor<DeviceContext, T, 2, Functor>(
      context.template device_context<DeviceContext>(), shuffled_x, *out, *dout,
      dx, {1});
  // transpose dX
  std::vector<int> origin_axis(x_dim.size());
  GetOriginDimFromShuffled(x_dim, dims, &origin_axis);
  Tensor dx_tmp;
  framework::TensorCopy(*dx, context.GetPlace(), &dx_tmp);
  dx_tmp.Resize(shuffled_dim);
  dx->Resize(x_dim);
  math::TransposeNormal<DeviceContext, T> trans;
  trans(context.template device_context<DeviceContext>(), dx_tmp, dx,
        origin_axis);
}
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200

template <typename DeviceContext, typename T, typename Functor>
struct ReduceKernelFunctor {
  const Tensor* input;
  Tensor* output;
  std::vector<int> dims;
  bool keep_dim;
  bool reduce_all;
  const framework::ExecutionContext& context;
  ReduceKernelFunctor(const Tensor* input, Tensor* output,
                      const std::vector<int>& dims, bool keep_dim,
                      bool reduce_all,
                      const framework::ExecutionContext& context)
      : input(input),
        output(output),
        dims(dims),
        keep_dim(keep_dim),
        reduce_all(reduce_all),
        context(context) {}

  template <typename OutT>
  void apply() const {
    output->mutable_data<OutT>(context.GetPlace());
    if (reduce_all) {
      // Flatten and reduce 1-D tensor
      auto x = EigenVector<OutT>::Flatten(*input);
      auto out = EigenScalar<OutT>::From(*output);
      auto& place =
          *context.template device_context<DeviceContext>().eigen_device();
      auto reduce_dim = Eigen::array<int, 1>({{0}});
      Functor functor;
      functor(place, &x, &out, reduce_dim);
    } else {
      int ndim = input->dims().size();
      int rdim = dims.size();
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221
      if (ndim > 6) {
        HandleLargeDim<DeviceContext, OutT, Functor>(context, input, output,
                                                     dims, keep_dim);
      } else {
        HANDLE_DIM(6, 5);
        HANDLE_DIM(6, 4);
        HANDLE_DIM(6, 3);
        HANDLE_DIM(6, 2);
        HANDLE_DIM(6, 1);
        HANDLE_DIM(5, 4);
        HANDLE_DIM(5, 3);
        HANDLE_DIM(5, 2);
        HANDLE_DIM(5, 1);
        HANDLE_DIM(4, 3);
        HANDLE_DIM(4, 2);
        HANDLE_DIM(4, 1);
        HANDLE_DIM(3, 2);
        HANDLE_DIM(3, 1);
        HANDLE_DIM(2, 1);
        HANDLE_DIM(1, 1);
      }
222 223 224
    }
  }
};
Q
QI JUN 已提交
225
template <typename DeviceContext, typename T, typename Functor>
Y
Yu Yang 已提交
226
class ReduceKernel : public framework::OpKernel<T> {
227 228 229 230 231 232 233 234 235
 public:
  void Compute(const framework::ExecutionContext& context) const override {
    bool reduce_all = context.Attr<bool>("reduce_all");
    auto* output = context.Output<Tensor>("Out");
    auto dims = context.Attr<std::vector<int>>("dim");
    bool keep_dim = context.Attr<bool>("keep_dim");
    int out_dtype = context.Attr<int>("out_dtype");
    framework::proto::VarType::Type cast_out_dtype;

236 237 238 239 240 241 242 243 244 245 246 247
    // The dims has full dim, set the reduce_all is True
    const auto& input_dim_size = context.Input<Tensor>("X")->dims().size();
    std::set<int> dims_set(dims.begin(), dims.end());
    bool full_dim = true;
    for (auto i = 0; i < input_dim_size; i++) {
      if (dims_set.find(i) == dims_set.end()) {
        full_dim = false;
        break;
      }
    }
    reduce_all = (reduce_all || full_dim);

248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
    if (out_dtype < 0) {
      auto* cast_input = context.Input<Tensor>("X");
      cast_out_dtype =
          static_cast<framework::proto::VarType::Type>(cast_input->type());
      framework::VisitDataType(
          cast_out_dtype,
          ReduceKernelFunctor<DeviceContext, T, Functor>(
              cast_input, output, dims, keep_dim, reduce_all, context));
    } else {
      Tensor tmp_tensor;
      cast_out_dtype = static_cast<framework::proto::VarType::Type>(out_dtype);
      auto* input = context.Input<Tensor>("X");

      tmp_tensor.Resize(input->dims());
      framework::VisitDataType(
          cast_out_dtype,
          CastOpFunctor<DeviceContext, T>(
              input, &tmp_tensor,
              context.template device_context<DeviceContext>()));
      framework::VisitDataType(
          cast_out_dtype,
          ReduceKernelFunctor<DeviceContext, T, Functor>(
              &tmp_tensor, output, dims, keep_dim, reduce_all, context));
    }
  }
};
template <typename DeviceContext, typename OutT, typename Functor>
class BoolReduceKernel : public framework::OpKernel<OutT> {
G
guosheng 已提交
276 277
 public:
  void Compute(const framework::ExecutionContext& context) const override {
278
    bool reduce_all = context.Attr<bool>("reduce_all");
279 280
    auto* input = context.Input<Tensor>("X");
    auto* output = context.Output<Tensor>("Out");
281
    output->mutable_data<OutT>(context.GetPlace());
282 283 284 285

    auto dims = context.Attr<std::vector<int>>("dim");
    bool keep_dim = context.Attr<bool>("keep_dim");

286 287 288 289 290 291 292 293 294 295 296 297
    // The dims has full dim, set the reduce_all is True
    const auto& input_dim_size = context.Input<Tensor>("X")->dims().size();
    std::set<int> dims_set(dims.begin(), dims.end());
    bool full_dim = true;
    for (auto i = 0; i < input_dim_size; i++) {
      if (dims_set.find(i) == dims_set.end()) {
        full_dim = false;
        break;
      }
    }
    reduce_all = (reduce_all || full_dim);

298 299
    if (reduce_all) {
      // Flatten and reduce 1-D tensor
300 301
      auto x = EigenVector<OutT>::Flatten(*input);
      auto out = EigenScalar<OutT>::From(*output);
302 303 304 305
      auto& place =
          *context.template device_context<DeviceContext>().eigen_device();
      auto reduce_dim = Eigen::array<int, 1>({{0}});
      Functor functor;
306
      functor(place, &x, &out, reduce_dim);
307
    } else {
308 309
      int ndim = input->dims().size();
      int rdim = dims.size();
310
      // comments for accelerating compiling temporarily.
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331
      if (ndim > 6) {
        HandleLargeDim<DeviceContext, OutT, Functor>(context, input, output,
                                                     dims, keep_dim);
      } else {
        HANDLE_DIM(6, 5);
        HANDLE_DIM(6, 4);
        HANDLE_DIM(6, 3);
        HANDLE_DIM(6, 2);
        HANDLE_DIM(6, 1);
        HANDLE_DIM(5, 4);
        HANDLE_DIM(5, 3);
        HANDLE_DIM(5, 2);
        HANDLE_DIM(5, 1);
        HANDLE_DIM(4, 3);
        HANDLE_DIM(4, 2);
        HANDLE_DIM(4, 1);
        HANDLE_DIM(3, 2);
        HANDLE_DIM(3, 1);
        HANDLE_DIM(2, 1);
        HANDLE_DIM(1, 1);
      }
G
guosheng 已提交
332 333 334
    }
  }
};
335

336 337
template <typename DeviceContext, typename T, typename Functor,
          bool kNoNeedBufferX = false, bool kNoNeedBufferY = false>
Y
Yu Yang 已提交
338
class ReduceGradKernel : public framework::OpKernel<T> {
G
guosheng 已提交
339
 public:
340 341
  void ComputeFromInput(const Tensor* input2,
                        const framework::ExecutionContext& context) const {
342
    bool reduce_all = context.Attr<bool>("reduce_all");
343 344 345
    auto dims = context.Attr<std::vector<int>>("dim");
    auto* input0 = context.Input<Tensor>("X");
    auto* input1 = context.Input<Tensor>("Out");
346

347 348 349
    auto* output = context.Output<Tensor>(framework::GradVarName("X"));
    output->mutable_data<T>(context.GetPlace());

350 351 352 353 354 355 356 357 358 359 360
    // The dims has full dim, set the reduce_all is True
    const auto& input_dim_size = context.Input<Tensor>("X")->dims().size();
    std::set<int> dims_set(dims.begin(), dims.end());
    bool full_dim = true;
    for (auto i = 0; i < input_dim_size; i++) {
      if (dims_set.find(i) == dims_set.end()) {
        full_dim = false;
        break;
      }
    }
    reduce_all = (reduce_all || full_dim);
361 362 363 364 365 366 367 368 369 370 371
    // NOTE: EigenTensor::From() uses tensor->data()
    // if op has NoNeedBufferVarsInferer, the corresponding kNoNeedBufferX or
    // kNoNeedBufferY should set true
    // and use fake var that has same dims.
    if (kNoNeedBufferX) {
      input0 = output;
    }
    if (kNoNeedBufferY) {
      input1 = input2;
    }

L
lvmengsi 已提交
372 373 374 375
    // NOTE(dengkaipeng): Out is unnecessary in some reduce kernel and
    // not be set as Input in grad Maker, use Out_grad to replace here
    if (!input1) input1 = input2;

376 377
    if (reduce_all) {
      auto x = EigenVector<T>::Flatten(*input0);
378 379
      auto x_reduce = EigenVector<T>::Flatten(*input1);
      auto x_reduce_grad = EigenVector<T>::Flatten(*input2);
380 381 382 383 384 385
      auto x_grad = EigenVector<T>::Flatten(*output);
      auto& place =
          *context.template device_context<DeviceContext>().eigen_device();
      auto broadcast_dim =
          Eigen::array<int, 1>({{static_cast<int>(input0->numel())}});
      Functor functor;
386
      functor(place, &x, &x_reduce, &x_grad, &x_reduce_grad, broadcast_dim,
387 388
              broadcast_dim[0]);
    } else {
389
      int rank = input0->dims().size();
390 391
      switch (rank) {
        case 1:
392 393 394
          ReduceGradFunctor<DeviceContext, T, 1, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
395 396
          break;
        case 2:
397 398 399
          ReduceGradFunctor<DeviceContext, T, 2, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
400 401
          break;
        case 3:
402 403 404
          ReduceGradFunctor<DeviceContext, T, 3, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
405 406
          break;
        case 4:
407 408 409
          ReduceGradFunctor<DeviceContext, T, 4, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
410 411
          break;
        case 5:
412 413 414
          ReduceGradFunctor<DeviceContext, T, 5, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
415 416
          break;
        case 6:
417 418 419
          ReduceGradFunctor<DeviceContext, T, 6, Functor>(
              context.template device_context<DeviceContext>(), *input0,
              *input1, *input2, output, dims);
420
          break;
421 422 423 424
        default:
          HandleLargeDimGrad<DeviceContext, T, Functor>(context, input0, input1,
                                                        input2, output, dims);
          break;
425
      }
G
guosheng 已提交
426 427
    }
  }
428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447

  void Compute(const framework::ExecutionContext& context) const override {
    int in_dtype = context.Attr<int>("in_dtype");
    if (in_dtype >= 0) {
      Tensor tmp_tensor;
      auto* pre_input = context.Input<Tensor>(framework::GradVarName("Out"));
      auto in_kernel_type =
          framework::OpKernelType(pre_input->type(), context.GetPlace());
      auto out_kernel_type = framework::OpKernelType(
          static_cast<framework::proto::VarType::Type>(in_dtype),
          context.GetPlace());
      framework::TransDataType(in_kernel_type, out_kernel_type, *pre_input,
                               &tmp_tensor);
      ComputeFromInput(&tmp_tensor, context);

    } else {
      auto* input2 = context.Input<Tensor>(framework::GradVarName("Out"));
      ComputeFromInput(input2, context);
    }
  }
448
};
G
guosheng 已提交
449

450 451 452
class ReduceOp : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;
G
guosheng 已提交
453

454
  void InferShape(framework::InferShapeContext* ctx) const override {
455 456
    OP_INOUT_CHECK(ctx->HasInput("X"), "Input", "X", "ReduceOp");
    OP_INOUT_CHECK(ctx->HasOutput("Out"), "Output", "Out", "ReduceOp");
457 458 459
    auto x_dims = ctx->GetInputDim("X");
    auto x_rank = x_dims.size();
    auto dims = ctx->Attrs().Get<std::vector<int>>("dim");
460 461 462 463 464 465
    PADDLE_ENFORCE_GT(dims.size(), 0,
                      platform::errors::InvalidArgument(
                          "The input dim dimensions of ReduceOp "
                          "should be greater than 0. But received the dim "
                          "dimesions of Reduce = %d.",
                          dims.size()));
466

467
    for (size_t i = 0; i < dims.size(); ++i) {
468
      PADDLE_ENFORCE_LT(dims[i], x_rank,
469 470 471 472 473
                        platform::errors::InvalidArgument(
                            "The reduce dim index %d should be in the "
                            "range [-dimension(X), dimension(X)] "
                            "which dimesion = %d. But received dim index = %d.",
                            i, x_rank, dims[i]));
474 475 476 477 478 479
      PADDLE_ENFORCE_GE(dims[i], -x_rank,
                        platform::errors::InvalidArgument(
                            "The reduce dim index %d should be in the "
                            "range [-dimension(X), dimension(X)] "
                            "which dimesion = %d. But received dim index = %d.",
                            i, x_rank, dims[i]));
480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
      if (dims[i] < 0) dims[i] = x_rank + dims[i];
    }
    sort(dims.begin(), dims.end());
    bool reduce_all = ctx->Attrs().Get<bool>("reduce_all");
    bool keep_dim = ctx->Attrs().Get<bool>("keep_dim");
    if (reduce_all) {
      if (keep_dim)
        ctx->SetOutputDim(
            "Out", framework::make_ddim(std::vector<int64_t>(x_rank, 1)));
      else
        ctx->SetOutputDim("Out", {1});
    } else {
      auto dims_vector = vectorize(x_dims);
      if (keep_dim) {
        for (size_t i = 0; i < dims.size(); ++i) {
          dims_vector[dims[i]] = 1;
        }
      } else {
        const int kDelFlag = -2;
        for (size_t i = 0; i < dims.size(); ++i) {
          dims_vector[dims[i]] = kDelFlag;
        }
        dims_vector.erase(
            remove(dims_vector.begin(), dims_vector.end(), kDelFlag),
            dims_vector.end());
      }
506 507 508
      if (!keep_dim && dims_vector.size() == 0) {
        dims_vector.push_back(1);
      }
509 510
      auto out_dims = framework::make_ddim(dims_vector);
      ctx->SetOutputDim("Out", out_dims);
511
      if (dims.size() > 0 && dims[0] != 0) {
512 513 514 515 516
        // Only pass LoD when not reducing on the first dim.
        ctx->ShareLoD("X", /*->*/ "Out");
      }
    }
  }
517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534

  framework::OpKernelType GetExpectedKernelType(
      const framework::ExecutionContext& ctx) const override {
    // choose cudnn kernel if the runtime supported.
    auto input_data_type = OperatorWithKernel::IndicateVarDataType(ctx, "X");

    if (ctx.Input<paddle::framework::LoDTensor>("X")->dims().size() > 5)
      return framework::OpKernelType(input_data_type, ctx.GetPlace());

#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

    if (input_data_type == framework::proto::VarType::FP16) {
F
furnace 已提交
535 536 537
      PADDLE_ENFORCE_EQ(platform::is_gpu_place(ctx.GetPlace()) ||
                            platform::is_npu_place(ctx.GetPlace()),
                        true,
538
                        platform::errors::InvalidArgument(
F
furnace 已提交
539
                            "float16 can only be used on GPU or NPU place"));
540 541 542
    }
    return framework::OpKernelType(input_data_type, ctx.GetPlace());
  }
543 544
};

G
Guo Sheng 已提交
545 546 547 548 549 550 551 552 553 554 555 556 557
class ReduceOpUseInputPlace : public ReduceOp {
 public:
  using ReduceOp::ReduceOp;

 protected:
  framework::OpKernelType GetExpectedKernelType(
      const framework::ExecutionContext& ctx) const override {
    framework::OpKernelType kt = OperatorWithKernel::GetExpectedKernelType(ctx);
    kt.place_ = ctx.Input<framework::LoDTensor>("X")->place();
    return kt;
  }
};

558 559 560
class ReduceGradOp : public framework::OperatorWithKernel {
 public:
  using framework::OperatorWithKernel::OperatorWithKernel;
W
whs 已提交
561

562
  void InferShape(framework::InferShapeContext* ctx) const override {
563 564 565
    OP_INOUT_CHECK(ctx->HasInput("X"), "Input", "X", "ReduceOp");
    OP_INOUT_CHECK(ctx->HasInput(framework::GradVarName("Out")), "Input",
                   "Out@GRAD", "ReduceOp");
566 567 568
    auto x_dims = ctx->GetInputDim("X");
    auto x_rank = x_dims.size();
    auto dims = ctx->Attrs().Get<std::vector<int>>("dim");
W
whs 已提交
569
    for (size_t i = 0; i < dims.size(); ++i) {
570
      PADDLE_ENFORCE_LT(dims[i], x_rank,
571 572 573 574 575
                        platform::errors::InvalidArgument(
                            "The reduce dim index %d should be in the "
                            "range [-dimension(X), dimension(X)], "
                            "which dimesion = %d. But received dim index = %d.",
                            i, x_rank, dims[i]));
W
whs 已提交
576
      if (dims[i] < 0) dims[i] = x_rank + dims[i];
577 578 579 580 581 582
    }
    sort(dims.begin(), dims.end());
    auto x_grad_name = framework::GradVarName("X");
    if (ctx->HasOutput(x_grad_name)) {
      ctx->SetOutputDim(x_grad_name, x_dims);
      ctx->ShareLoD("X", /*->*/ x_grad_name);
W
whs 已提交
583
    }
584
  }
585 586 587 588

 protected:
  framework::OpKernelType GetExpectedKernelType(
      const framework::ExecutionContext& ctx) const override {
J
jakpiase 已提交
589 590 591 592 593
    int in_dtype = ctx.Attr<int>("in_dtype");
    auto input_data_type =
        (in_dtype >= 0) ? static_cast<framework::proto::VarType::Type>(in_dtype)
                        : OperatorWithKernel::IndicateVarDataType(
                              ctx, framework::GradVarName("Out"));
594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610
#ifdef PADDLE_WITH_MKLDNN
    auto CanMKLDNNReduceGradBeUsed = [&]() {
      auto dx_dims = ctx.Input<Tensor>("X")->dims();

      if (dx_dims.size() > 5) return false;  // max 5D tensor is supported

      return true;
    };
    if (this->CanMKLDNNBeUsed(ctx, input_data_type) &&
        CanMKLDNNReduceGradBeUsed()) {
      return framework::OpKernelType(input_data_type, ctx.GetPlace(),
                                     framework::DataLayout::kMKLDNN,
                                     framework::LibraryType::kMKLDNN);
    }
#endif

    return framework::OpKernelType(input_data_type, ctx.GetPlace());
611
  }
612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
};

class ReduceOpMaker : public framework::OpProtoAndCheckerMaker {
 public:
  void Make() final {
    AddInput("X",
             "(Tensor) The input tensor. Tensors with rank at most 6 are "
             "supported.");
    AddOutput("Out", "(Tensor) The result tensor.");
    AddAttr<std::vector<int>>(
        "dim",
        "(list<int>, default {0}) The dimensions to reduce. "
        "Must be in the range [-rank(input), rank(input)). "
        "If `dim[i] < 0`, the dims[i] to reduce is `rank + dims[i]`. "
        "Note that reducing on the first dim will make the LoD info lost.")
        .SetDefault({0});
    AddAttr<bool>("keep_dim",
                  "(bool, default false) "
                  "If true, retain the reduced dimension with length 1.")
        .SetDefault(false);
    AddAttr<bool>("reduce_all",
                  "(bool, default false) "
                  "If true, output a scalar reduced along all dimensions.")
        .SetDefault(false);
636 637 638 639 640 641 642 643 644 645
    AddAttr<int>("in_dtype",
                 "(int, default -1)"
                 "The dtype of input, default value is -1, the user could not "
                 "set this value.")
        .SetDefault(-1);
    AddAttr<int>(
        "out_dtype",
        "(int, default -1)"
        "The dtype of output, default value is -1, the dtype is same as intput")
        .SetDefault(-1);
646 647 648
    AddAttr<bool>("use_mkldnn",
                  "(bool, default false) Only used in mkldnn kernel")
        .SetDefault(false);
649 650
    AddComment(string::Sprintf(R"DOC(
%s Operator.
W
whs 已提交
651

652 653 654
This operator computes the %s of input tensor along the given dimension.
The result tensor has 1 fewer dimension than the input unless keep_dim is true.
If reduce_all is true, just reduce along all dimensions and output a scalar.
W
whs 已提交
655

656 657
)DOC",
                               GetOpType(), GetName()));
G
guosheng 已提交
658
  }
659 660 661 662

 protected:
  virtual std::string GetName() const = 0;
  virtual std::string GetOpType() const = 0;
G
guosheng 已提交
663 664
};

665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691
#if defined(__HIPCC__) || defined(__NVCC__)
template <typename T, template <typename, typename> class ReduceOp>
class ReduceCudaKernel : public framework::OpKernel<T> {
 public:
  void Compute(const framework::ExecutionContext& context) const override {
    bool reduce_all = context.Attr<bool>("reduce_all");
    const Tensor* input = context.Input<Tensor>("X");
    Tensor* output = context.Output<Tensor>("Out");
    auto out_dtype = context.Attr<int>("out_dtype");
    std::vector<int> dims = context.Attr<std::vector<int>>("dim");

    std::vector<int> reduce_dims =
        GetReduceDim(dims, input->dims().size(), reduce_all);

    gpuStream_t stream = context.cuda_device_context().stream();
    if (out_dtype >= 0) {
      framework::VisitDataTypeSmall(
          static_cast<framework::proto::VarType::Type>(out_dtype),
          TensorReduceFunc<T, ReduceOp>(*input, output, reduce_dims, stream));
    } else {
      TensorReduceFunctorImpl<T, T, ReduceOp>(*input, output, reduce_dims,
                                              stream);
    }
  }
};
#endif

G
guosheng 已提交
692 693
}  // namespace operators
}  // namespace paddle
694

695 696
namespace ops = paddle::operators;

H
hong 已提交
697 698 699 700 701 702 703 704 705 706 707 708 709 710
#define REGISTER_REDUCE_OP(op_name)                                           \
  class __##op_name##Maker__ : public ops::ReduceOpMaker {                    \
   protected:                                                                 \
    virtual std::string GetName() const { return #op_name; }                  \
    virtual std::string GetOpType() const { return "Reduce " #op_name; }      \
  };                                                                          \
  REGISTER_OPERATOR(                                                          \
      op_name, ops::ReduceOp, __##op_name##Maker__,                           \
      paddle::framework::DefaultGradOpMaker<paddle::framework::OpDesc, true>, \
      paddle::framework::DefaultGradOpMaker<paddle::imperative::OpBase,       \
                                            true>);                           \
  REGISTER_OPERATOR(op_name##_grad, ops::ReduceGradOp)

#define REGISTER_REDUCE_OP_WITHOUT_GRAD(op_name, ...)                    \
711 712 713 714 715
  class __##op_name##Maker__ : public ops::ReduceOpMaker {               \
   protected:                                                            \
    virtual std::string GetName() const { return #op_name; }             \
    virtual std::string GetOpType() const { return "Reduce " #op_name; } \
  };                                                                     \
H
hong 已提交
716 717 718 719
  REGISTER_OPERATOR(                                                     \
      op_name, ops::ReduceOp##__VA_ARGS__, __##op_name##Maker__,         \
      paddle::framework::EmptyGradOpMaker<paddle::framework::OpDesc>,    \
      paddle::framework::EmptyGradOpMaker<paddle::imperative::OpBase>);