1. 18 1月, 2018 3 次提交
  2. 12 12月, 2017 1 次提交
    • Q
      Refine device context (#6433) · 61ec0b95
      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`
      61ec0b95
  3. 18 10月, 2017 1 次提交
    • M
      MatMul operator (#4856) · 16489827
      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.
      16489827