matmul_v2_op.cu 2.0 KB
Newer Older
S
ShenLiang 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
/* Copyright (c) 2020 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. */

#include "paddle/fluid/operators/matmul_v2_op.h"

namespace ops = paddle::operators;
namespace plf = paddle::platform;

S
ShenLiang 已提交
20 21 22
REGISTER_OP_CUDA_KERNEL(
    matmul_v2, ops::MatMulV2Kernel<plf::CUDADeviceContext, float>,
    ops::MatMulV2Kernel<plf::CUDADeviceContext, double>,
23
    ops::MatMulV2Kernel<plf::CUDADeviceContext, plf::float16>,
24 25
    ops::MatMulV2Kernel<plf::CUDADeviceContext, plf::complex<float>>,
    ops::MatMulV2Kernel<plf::CUDADeviceContext, plf::complex<double>>);
S
ShenLiang 已提交
26 27 28

REGISTER_OP_CUDA_KERNEL(
    matmul_v2_grad, ops::MatMulV2GradKernel<plf::CUDADeviceContext, float>,
S
ShenLiang 已提交
29
    ops::MatMulV2GradKernel<plf::CUDADeviceContext, double>,
30
    ops::MatMulV2GradKernel<plf::CUDADeviceContext, plf::float16>,
31 32
    ops::MatMulV2GradKernel<plf::CUDADeviceContext, plf::complex<float>>,
    ops::MatMulV2GradKernel<plf::CUDADeviceContext, plf::complex<double>>);
W
wawltor 已提交
33 34 35 36 37 38 39 40 41 42

REGISTER_OP_CUDA_KERNEL(
    matmul_v2_grad_grad,
    ops::MatMulV2DoubleGradKernel<paddle::platform::CUDADeviceContext, float>,
    ops::MatMulV2DoubleGradKernel<paddle::platform::CUDADeviceContext, double>,
    ops::MatMulV2DoubleGradKernel<plf::CUDADeviceContext, plf::float16>,
    ops::MatMulV2DoubleGradKernel<paddle::platform::CUDADeviceContext,
                                  paddle::platform::complex<float>>,
    ops::MatMulV2DoubleGradKernel<paddle::platform::CUDADeviceContext,
                                  paddle::platform::complex<double>>);