1. 13 10月, 2019 2 次提交
    • G
      Cherry pick seq2seq api from #19820 (#20555) · 36ffc8b1
      Guo Sheng 提交于
      * Add seq2seq api related code (#19820)
      
      * fix expand bug  (#20340)
      
      * fix expand bug test=develop
      
      * fix style test=develop
      
      * fix style test=develop
      
      * fix style test=develop
      
      * fix style test=develop
      
      * Fix the assign data check (#20564)
      
      * Fix the assign data check. test=develop
      
      * Fix test_assign_op.py. test=develop
      
      * Update API.spec
      36ffc8b1
    • L
      [cherry-pick #20521] (#20590) · 89e64193
      liym27 提交于
      1. fill_constant support Tensor;
      2. fix bug in backward.py: using fill_constant instead of fill_constant_batch_size_like
      3. fix bug in ExpandGradOp.
      test=release/1.6 (#20521)
      89e64193
  2. 17 9月, 2019 1 次提交
    • L
      fix expand op: (#19302) · e9e3c087
      liym27 提交于
      1. add tensor support for argument expand_times in expand op;
      2. add support parameter inference when argument expand_times is a list containing integer and tensor variable;
      
      improve expand op according to reviews:
      1. add doc of ExpandTimes in expand_op.cc;
      2. improve the test of test_api.
      
      add stop_gradient=True when attr(expand_times) is tensor Variable, change code examples.
      test=develop,test=document_preview
      e9e3c087
  3. 10 9月, 2019 1 次提交
  4. 04 6月, 2019 1 次提交
    • H
      expand op supprt tensor attribute (#17773) · dfec6762
      Hongyu Liu 提交于
      * expand support tensor attribute; test=develop
      
      * fix bug ; test=develop
      
      * fix uni test bug; test=develop
      
      * fix copy bug; test=develop
      
      * refine expand_times default value; test=develop
      dfec6762
  5. 14 3月, 2019 2 次提交
  6. 12 3月, 2019 1 次提交
  7. 11 3月, 2019 1 次提交
  8. 13 2月, 2019 1 次提交
  9. 14 1月, 2019 1 次提交
  10. 11 1月, 2019 1 次提交
  11. 15 11月, 2018 4 次提交
  12. 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
  13. 20 4月, 2018 1 次提交
  14. 19 4月, 2018 2 次提交
  15. 17 4月, 2018 1 次提交
  16. 12 2月, 2018 1 次提交
  17. 10 2月, 2018 2 次提交
  18. 16 1月, 2018 1 次提交
  19. 20 12月, 2017 1 次提交
  20. 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
  21. 10 11月, 2017 2 次提交
  22. 09 11月, 2017 1 次提交
  23. 13 9月, 2017 2 次提交