“3b094cd21a451662d66660d2c96993470190f400”上不存在“tests/git@gitcode.net:taosdata/tdengine.git”
clip_by_norm_op.h 1.9 KB
Newer Older
W
wwhu 已提交
1 2
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.

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
W
wwhu 已提交
6

L
Luo Tao 已提交
7
    http://www.apache.org/licenses/LICENSE-2.0
W
wwhu 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. */
W
wwhu 已提交
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

#pragma once

#include "paddle/framework/eigen.h"
#include "paddle/framework/op_registry.h"
#include "paddle/platform/transform.h"

namespace paddle {
namespace operators {

using Tensor = framework::Tensor;
template <typename T, int MajorType = Eigen::RowMajor,
          typename IndexType = Eigen::DenseIndex>
using EigenVector = framework::EigenVector<T, MajorType, IndexType>;

Q
QI JUN 已提交
29
template <typename DeviceContext, typename T>
W
wwhu 已提交
30 31 32 33 34 35 36 37 38 39 40
class ClipByNormKernel : public framework::OpKernel<T> {
 public:
  void Compute(const framework::ExecutionContext& context) const override {
    auto max_norm = context.Attr<T>("max_norm");
    auto* input = context.Input<Tensor>("X");
    auto* output = context.Output<Tensor>("Out");
    output->mutable_data<T>(context.GetPlace());

    auto x = EigenVector<T>::Flatten(*input);
    auto out = EigenVector<T>::Flatten(*output);
    auto x_norm = x.square().sum().sqrt();
Q
QI JUN 已提交
41 42
    auto& place =
        *context.template device_context<DeviceContext>().eigen_device();
W
wwhu 已提交
43 44 45 46 47 48 49 50 51 52 53

    auto temp = (x_norm <= max_norm).template cast<T>().eval();
    auto scaling = temp + (static_cast<T>(1) - temp) * max_norm / x_norm;
    Eigen::array<int, 1> one_dim{{1}};
    Eigen::DSizes<int, 1> m_dsize(input->numel());
    out.device(place) = x * scaling.reshape(one_dim).broadcast(m_dsize);
  }
};

}  // namespace operators
}  // namespace paddle