1. 01 10月, 2019 1 次提交
  2. 27 9月, 2019 1 次提交
  3. 18 9月, 2019 1 次提交
    • L
      Update elementwise double grad to save gpu memory (#19509) · 982e61f5
      Leo Chen 提交于
      * update elementwise double grad to save gpu memory, test=develop
      
      * update elementwise_mul/div_grad_grad to save memory, test=develop
      
      * remove eval function in eigen statement to save memory, test=develop
      
      * add unittest for elementwise_div_grad_grad without dout, test=develop
      
      * add unittest for elementwise_add_grad_grad without ddx, test=develop
      
      * add float16 cuda kernel for elementwise double grad op, test=develop
      982e61f5
  4. 13 5月, 2019 1 次提交
    • K
      add double grad for elementwise_mul op (#17255) · 8bae8590
      Kaipeng Deng 提交于
      * add double grad for elementwise_mul. test=develop
      
      * remove comment. test=develop
      
      * fix grad sum. test=develop
      
      * fix for axis expand. test=develop
      
      * add test for axis expand. test=develop
      8bae8590
  5. 16 11月, 2018 1 次提交
    • W
      Refine operator cmake (#14413) · a2d9b344
      Wu Yi 提交于
      * wip simplify operator framework
      
      * wip
      
      * wip
      
      * done test=develop
      
      * clean test=develop
      
      * fix test=develop
      
      * fix deps test=develop
      
      * fix cpu build test=develop
      
      * fix tensorrt build test=develop
      
      * fix tests test=develop
      
      * fix test=develop
      
      * fix cpu build test=develop
      a2d9b344
  6. 17 9月, 2018 2 次提交
  7. 15 6月, 2018 1 次提交
  8. 08 5月, 2018 1 次提交
    • Y
      Clean OpProtoAndCheckerMaker · 0e78cb69
      Yu Yang 提交于
      Do not use ctor
      
      * Reduce line of codes.
      * We can use virtual function for Maker now.
      * The implementation does not care what maker holds, it is easier to
      refactor later.
      0e78cb69
  9. 19 4月, 2018 1 次提交
  10. 17 4月, 2018 1 次提交
  11. 12 2月, 2018 1 次提交
  12. 10 2月, 2018 2 次提交
  13. 11 1月, 2018 1 次提交
  14. 26 12月, 2017 1 次提交
  15. 20 12月, 2017 1 次提交
  16. 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
  17. 10 11月, 2017 1 次提交
  18. 04 11月, 2017 1 次提交
  19. 28 9月, 2017 1 次提交
  20. 27 9月, 2017 1 次提交
  21. 22 9月, 2017 1 次提交
  22. 21 9月, 2017 1 次提交
  23. 20 9月, 2017 1 次提交
  24. 15 9月, 2017 1 次提交
  25. 14 9月, 2017 2 次提交
  26. 13 9月, 2017 1 次提交