- 14 4月, 2018 1 次提交
-
-
由 Abhinav Arora 提交于
* Fix CPPLint errors in operators * Fix cast in softmax * Fix softmax_mkldnn * Fix send_recv_op_test * Send_recv * Fix softmax mkldnn
-
- 12 2月, 2018 2 次提交
-
-
由 qingqing01 提交于
-
由 typhoonzero 提交于
-
- 10 2月, 2018 2 次提交
- 09 2月, 2018 1 次提交
-
-
由 Yancey1989 提交于
-
- 08 2月, 2018 1 次提交
-
-
由 Yancey1989 提交于
-
- 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 9月, 2017 2 次提交
- 18 9月, 2017 1 次提交
-
-
由 Yancey 提交于
Split Op CPU Kernel
-
- 11 9月, 2017 1 次提交
-
-
由 Yancey 提交于
add concat op with CPU kernel
-