- 18 1月, 2018 3 次提交
-
-
由 chengduoZH 提交于
-
由 chengduoZH 提交于
-
由 chengduoZH 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`
-
- 18 10月, 2017 1 次提交
-
-
由 Markus Kliegl 提交于
* initial matmul operator Similar to np.matmul, but also has transpose_X and transpose_Y flags, and only supports tensors from rank 1 to 3 inclusive. For GPU, uses cublas?gemmStridedBatched. For CPU, uses cblas_?gemm_batch if available via MKL; otherwise a simple serial implementation that loops over the batch dimension is employed for now.
-