transpose_op.cu.cc 1.8 KB
Newer Older
1
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
X
xzl 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
X
xzl 已提交
6

L
Luo Tao 已提交
7
    http://www.apache.org/licenses/LICENSE-2.0
X
xzl 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. */
X
xzl 已提交
14

Y
Yi Wang 已提交
15
#include "paddle/fluid/operators/transpose_op.h"
X
xzl 已提交
16 17

namespace ops = paddle::operators;
18 19
namespace plat = paddle::platform;

Q
QI JUN 已提交
20
REGISTER_OP_CUDA_KERNEL(
P
phlrain 已提交
21
    transpose, ops::TransposeKernel<paddle::platform::CUDADeviceContext, float>,
22 23
    ops::TransposeKernel<paddle::platform::CUDADeviceContext, double>,
    ops::TransposeKernel<paddle::platform::CUDADeviceContext, plat::float16>);
Q
QI JUN 已提交
24
REGISTER_OP_CUDA_KERNEL(
25
    transpose_grad,
P
phlrain 已提交
26
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext, float>,
27 28 29
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext, double>,
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext,
                             plat::float16>);
30 31 32

REGISTER_OP_CUDA_KERNEL(
    transpose2,
P
phlrain 已提交
33
    ops::TransposeKernel<paddle::platform::CUDADeviceContext, float>,
34 35
    ops::TransposeKernel<paddle::platform::CUDADeviceContext, double>,
    ops::TransposeKernel<paddle::platform::CUDADeviceContext, plat::float16>);
36 37
REGISTER_OP_CUDA_KERNEL(
    transpose2_grad,
P
phlrain 已提交
38
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext, float>,
39 40 41
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext, double>,
    ops::TransposeGradKernel<paddle::platform::CUDADeviceContext,
                             plat::float16>);