- 18 3月, 2018 1 次提交
-
-
由 Kexin Zhao 提交于
-
- 17 3月, 2018 1 次提交
-
-
由 Kexin Zhao 提交于
-
- 12 2月, 2018 1 次提交
-
-
由 qingqing01 提交于
-
- 10 2月, 2018 2 次提交
- 25 12月, 2017 1 次提交
-
-
由 dzhwinter 提交于
-
- 22 12月, 2017 1 次提交
-
-
由 QI JUN 提交于
* add data layout * fix ci
-
- 12 12月, 2017 1 次提交
-
-
由 QI JUN 提交于
There are mainly following fixes: - take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place` - remove `eigen_device` interface in base class `DeviceContext` - remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext` - remove unused `platform::EigenDeviceConverter` - rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL` - rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`
-
- 28 11月, 2017 1 次提交
-
-
由 Qiao Longfei 提交于
* batch norm support matrix input * update gpu code * format code
-
- 11 11月, 2017 1 次提交
-
-
由 dangqingqing 提交于
-
- 28 10月, 2017 2 次提交
- 26 10月, 2017 1 次提交
-
-
由 Qiao Longfei 提交于
* init cudnn batch norm op * rename batch_norm_cudnn_op.cc batch_norm_op.cu * correct name style * add ExtractNCWHD, simplify code * fix ExtractNCWHD * use CUDNN_ENFORCE instead of PADDLE_ENFORCE
-