1. 13 3月, 2022 1 次提交
  2. 08 3月, 2022 1 次提交
  3. 14 4月, 2021 1 次提交
    • Z
      fix matrix_inverse_op with rocm (#32128) · 995b5f2c
      zhulei 提交于
      * fix matrix_inverse_op with rocm
      
      * fix matrix_inverse_op with rocm
      
      * fix matrix_inverse_op with rocm
      
      * fix matrix_inverse_op with rocm
      995b5f2c
  4. 27 4月, 2020 1 次提交
  5. 26 11月, 2019 1 次提交
    • G
      Add fc padding to improve mkl GEMM's performance when N and K are multiple of 128. (#20972) · 234060f8
      GaoWei8 提交于
      * Add fc padding to solve mkl performance
      test=develop
      
      * fix gpu pass and error information
      test=develop
      
      * fix fc_fuse_pass_test
      test=develop
      
      * fix error information
      test=develop
      
      * fix error information
      test=develop
      
      * fix name and add fc op padding test
      test=develop
      
      * fix attributes
      test=develop
      
      * optimize fc padding
      test=develop
      
      * fix test
      test=develop
      234060f8
  6. 11 9月, 2019 1 次提交
    • Y
      Implement the GPU kernel of fc operator (#19687) · a65c728e
      Yiqun Liu 提交于
      * Refine the codes related to fc op.
      
      * Add GPU implementation for fc functor.
      
      * Apply fc_fuse_pass in GPU inference.
      test=develop
      
      * Change the cmake for fc op.
      
      * Change PADDLE_ENFORCE to PADDLE_ENFORCE_EQ.
      
      * Add an attribute to set the activation type in fc_op.
      
      * Enhance the unittest of fc_op.
      test=develop
      
      * Remove the declaration of FCOpGrad back to the header file.
      test=develop
      
      * Set default value for newly added arguments in test_fc_op.
      test=develop
      a65c728e
  7. 19 2月, 2019 1 次提交
  8. 30 1月, 2019 1 次提交
  9. 12 2月, 2018 1 次提交
  10. 10 2月, 2018 2 次提交
  11. 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
  12. 26 10月, 2017 1 次提交
  13. 23 10月, 2017 1 次提交
  14. 28 9月, 2017 1 次提交
  15. 27 9月, 2017 1 次提交
  16. 26 9月, 2017 1 次提交
  17. 22 9月, 2017 1 次提交
  18. 13 9月, 2017 1 次提交
  19. 12 9月, 2017 1 次提交