merge_ids_op.h 3.2 KB
Newer Older
Q
qiaolongfei 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 27 28 29 30 31 32
/* Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserved.

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. */

#pragma once

#include <vector>
#include "paddle/fluid/framework/op_registry.h"
#include "paddle/fluid/framework/tensor_util.h"
#include "paddle/fluid/operators/math/selected_rows_functor.h"

namespace paddle {
namespace operators {

template <typename DeviceContext, typename T>
class MergeIdsOpKernel : public framework::OpKernel<T> {
 public:
  void Compute(const framework::ExecutionContext &ctx) const override {
    auto place = ctx.GetPlace();
    if (!platform::is_cpu_place(place)) {
      PADDLE_THROW("MergeIds do not support GPU kernel");
    }
Q
qiaolongfei 已提交
33
    VLOG(3) << "run in MergeIdsOpKernel";
Q
qiaolongfei 已提交
34 35 36 37 38 39 40

    const auto *ids_var = ctx.InputVar("Ids");
    PADDLE_ENFORCE(ids_var->IsType<framework::LoDTensor>(),
                   "only support to merge Ids of LoDTensor");

    const auto &ids_tensor = ids_var->Get<framework::LoDTensor>();
    const auto &ids_dims = ids_tensor.dims();
Q
qiaolongfei 已提交
41
    const int64_t *ids = ids_tensor.data<int64_t>();
Q
qiaolongfei 已提交
42 43 44 45 46 47 48 49

    auto x_tensors = ctx.MultiInput<framework::LoDTensor>("X");

    auto *out = ctx.Output<framework::LoDTensor>("Out");

    int batch_size = 0;
    int embedding_size = 0;
    for (auto &input : x_tensors) {
Q
qiaolongfei 已提交
50
      if (framework::product(input->dims()) != 0) {
Q
qiaolongfei 已提交
51 52 53
        if (embedding_size == 0) {
          embedding_size = input->dims()[1];
        }
Q
qiaolongfei 已提交
54 55 56 57
        PADDLE_ENFORCE_EQ(embedding_size, input->dims()[1],
                          "embedding size of all input should be the same");
        batch_size += input->dims()[0];
      }
Q
qiaolongfei 已提交
58 59 60
    }
    PADDLE_ENFORCE_EQ(
        batch_size, ids_dims[0],
Q
qiaolongfei 已提交
61
        "the batch size of ids and merged embedding value should be the same");
Q
qiaolongfei 已提交
62 63 64 65 66

    const size_t shard_num = x_tensors.size();

    if (shard_num == 1) {
      VLOG(3) << "only one shard, we can copy the data directly";
Q
qiaolongfei 已提交
67
      TensorCopy(*x_tensors[0], place, out);
Q
qiaolongfei 已提交
68 69
    } else {
      std::vector<int> in_indexs(shard_num, 0);
Q
qiaolongfei 已提交
70 71
      auto *out_data = out->mutable_data<T>(
          framework::make_ddim({batch_size, embedding_size}), place);
Q
qiaolongfei 已提交
72 73
      // copy data from ins[shard_num] to out.
      for (int i = 0; i < ids_dims[0]; ++i) {
Q
qiaolongfei 已提交
74
        int64_t id = ids[i];
Q
qiaolongfei 已提交
75 76 77 78 79 80 81
        size_t shard_id = static_cast<size_t>(id) % shard_num;
        int index = in_indexs[shard_id];
        memcpy(out_data + embedding_size * i,
               x_tensors[shard_id]->data<T>() + index * embedding_size,
               sizeof(T) * embedding_size);
        in_indexs[shard_id] += 1;
      }
Q
qiaolongfei 已提交
82

Q
qiaolongfei 已提交
83
      for (size_t i = 0; i < shard_num; ++i) {
Q
qiaolongfei 已提交
84 85 86
        PADDLE_ENFORCE_EQ(in_indexs[i], x_tensors[i]->dims()[0],
                          "after merge, all data in x_tensor should be used");
      }
Q
qiaolongfei 已提交
87 88 89 90 91 92
    }
  }
};

}  // namespace operators
}  // namespace paddle