1. 08 9月, 2020 1 次提交
  2. 11 7月, 2020 1 次提交
  3. 12 5月, 2020 1 次提交
  4. 11 5月, 2020 1 次提交
    • C
      Add macro BOOST_GET to enrich the error information of boost :: get (#24175) · aa0f254f
      Chen Weihang 提交于
      * add new macro BOOST_GET_SAFELY & unittests, test=develop
      
      * add different macro type, test=develop
      
      * fix get macro type in executor, test=develop
      
      * four macro part change backup
      
      * using one macro for all case, test=develop
      
      * revert attribute change, test=develop
      
      * change to three func to solve gcc4.8 bug, test=develop
      
      * polish some details, test=develop
      aa0f254f
  5. 11 4月, 2020 1 次提交
  6. 10 4月, 2020 1 次提交
  7. 10 10月, 2019 1 次提交
  8. 11 9月, 2019 1 次提交
    • H
      Replace TemporaryAllocator by CUDADeviceContextAllocator (#18989) · 12542320
      Huihuang Zheng 提交于
      TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory.
      
      We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton.
      
      Also added data_feed_proto to operator to fix CI in CPU compilation
      12542320
  9. 23 1月, 2019 1 次提交
  10. 18 10月, 2018 1 次提交
  11. 16 10月, 2018 2 次提交
  12. 15 8月, 2018 1 次提交
  13. 01 6月, 2018 1 次提交
  14. 12 2月, 2018 1 次提交
  15. 10 2月, 2018 2 次提交
  16. 22 12月, 2017 1 次提交
  17. 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
  18. 25 10月, 2017 1 次提交
    • Q
      CPU Batch Norm Op (#4964) · ee998a9c
      Qiao Longfei 提交于
      * init batch norm op
      
      * prepare input output
      
      * compute mean_out var_out save_mean save_var on CPU
      
      * active is test
      
      * use eigen to do computation
      
      * complete batch norm forward
      
      * set default momentum to 0.9
      
      * add batch norm grad op in CPU
      
      * add tensor_format and NHWC support, add python test
      
      * add test training
      
      * add batch norm gradient test
      
      * improve comment, fix foward Python UnitTest
      
      * add gradient test
      
      * fix eigen warning
      
      * follow name style
      
      * fix a bug
      
      * change float to T
      
      * add simple forward test
      
      * test with different place
      
      * add backward test
      
      * refine python test
      
      * remove old python test code
      
      * code clean
      
      * follow code style
      
      * update comment
      ee998a9c
  19. 10 10月, 2017 1 次提交
  20. 28 9月, 2017 1 次提交
  21. 20 9月, 2017 1 次提交
  22. 23 8月, 2017 1 次提交
  23. 11 8月, 2017 1 次提交
  24. 08 8月, 2017 1 次提交
  25. 07 8月, 2017 1 次提交
  26. 05 8月, 2017 1 次提交
  27. 02 8月, 2017 1 次提交
  28. 01 8月, 2017 1 次提交
  29. 26 7月, 2017 1 次提交
  30. 25 7月, 2017 1 次提交
  31. 19 7月, 2017 2 次提交
  32. 17 7月, 2017 3 次提交
  33. 14 7月, 2017 1 次提交
  34. 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