1. 20 9月, 2019 1 次提交
  2. 10 6月, 2019 1 次提交
    • Y
      Enable seq_pool op to accept len 0 input (#17284) · 33d1e565
      Yibing Liu 提交于
      * Enable seq_pool op to accept len 0 input
      
      test=develop
      
      * Update sequence_pool's api
      
      test=develop
      
      * Add more unittest cases for seq_pool op
      
      test=develop
      
      * Remove legacy comments
      
      test=develop
      
      * Don't use template in op maker
      
      test=develop
      33d1e565
  3. 29 10月, 2018 1 次提交
    • J
      Rebase of seqpool-max optimization · 458b16f4
      Jacek Czaja 提交于
      test=develop
      
      - Added rough profiling
      
      - Profiled maxpool itself
      
      - First draft of max seqpool optimization (is_test added)
      
      - Added unit tests to seqpool
      
      - Cosmetic fixes
      
      - Fix to UT of Seq pool
      
      Disabled grad checking for sequence max pool when is_test is set to True
      
      -Cosmetic fix to comment
      
      test=develop
      
      - Fix to GPU build
      
      test=develop
      
      - yet another GPU fix for sequence max pool
      
      - Fix to comment
      
      test=develop
      
      - Change to API of sequence_pool
      
      test=develop
      
      - Yet another API spec change
      
      test=develop
      458b16f4
  4. 27 4月, 2018 1 次提交
  5. 29 3月, 2018 1 次提交
    • D
      Speed/sequence op1 (#9217) · 8425c2c8
      dzhwinter 提交于
      * "add functors"
      
      * "remove old code"
      
      * "fix"
      
      * "fix ci"
      
      * "add details"
      
      * "fix ci"
      
      * "fix ci"
      
      * "fix ci"
      
      * "fix ci"
      
      * "remove unused code"
      8425c2c8
  6. 12 2月, 2018 1 次提交
  7. 10 2月, 2018 2 次提交
  8. 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
  9. 03 11月, 2017 1 次提交
  10. 26 10月, 2017 1 次提交
  11. 10 10月, 2017 1 次提交
  12. 28 9月, 2017 1 次提交
  13. 20 9月, 2017 1 次提交
  14. 23 8月, 2017 1 次提交
  15. 11 8月, 2017 1 次提交
  16. 08 8月, 2017 1 次提交
  17. 07 8月, 2017 1 次提交
  18. 05 8月, 2017 1 次提交
  19. 02 8月, 2017 1 次提交
  20. 01 8月, 2017 1 次提交
  21. 26 7月, 2017 1 次提交
  22. 25 7月, 2017 1 次提交
  23. 19 7月, 2017 2 次提交
  24. 17 7月, 2017 3 次提交
  25. 14 7月, 2017 1 次提交
  26. 13 7月, 2017 2 次提交
    • Y
      Follow comments · 79b70c2d
      Yu Yang 提交于
      * Convert `op` --> `operators`
      * Remove AddType in OpProtoMaker, because type is part of registry.
      * Rename CPU_OR_GPU --> DEVICE_TYPE in registry macro.
      79b70c2d
    • Y
      Add a sample op, `add_op` · a0aaafe9
      Yu Yang 提交于
      * Refine register methods, make Op can get rid of whole-archieve
      * `USE_OP` before a op is used.
      * Add unittest for add_op.
      a0aaafe9