1. 18 2月, 2022 1 次提交
    • Z
      [AMP] support GPU BF16 amp for dygraph (#39029) · 7d6d3848
      zhangbo9674 提交于
      * support dtype param for auto_cast
      
      * add amp_dtype for tracer
      
      * add unsupported bf16 list
      
      * support bf16 amp for O2
      
      * refine python interface for bfloat16
      
      * refine code
      
      * refine code
      
      * refine unittest
      
      * refine code
      
      * refine code
      
      * add bf16 o1
      
      * refine code by comment
      
      * add gradient accumulator
      
      * add recompute
      7d6d3848
  2. 16 2月, 2022 1 次提交
    • J
      EagerTensor to EagerVariable (#39447) · 831fd86e
      Jiabin Yang 提交于
      * merge legacy to fluid
      
      * Remove legacy code
      
      * Remove legacy code
      
      * Remove DataType test
      
      * Using Tensor directly instead of using EagerTensor
      
      * support gradient_accumulation
      
      * make test_imperative_lod_tensor_to_selected_rows longer
      
      * make test_imperative_lod_tensor_to_selected_rows longer
      
      * refine code
      
      * Rename all EagerTensor to Tensor
      
      * Rename some EagerTensor to Tensor
      
      * rename EagerTensor to EagerVariable
      
      * add more test
      
      * merge develop and refine code
      831fd86e
  3. 24 12月, 2021 1 次提交
    • J
      Support test imperative basic in eager (#38313) · d48f7c89
      Jiabin Yang 提交于
      * Rearranged Eager AutoCodeGen directory structure
      
      * Removed USE_OP in Eager AutoCodeGen
      
      * Enabled generation for Operators without Grad/Inputs/Outputs
      
      * Resolved operators without input
      
      * Fixed merge conflicts
      
      * Enabled Eager AutoCodeGen for 10+ more operators
      
      * Refactored Eager AutoCodeGen with more organized helper objects
      
      * Enabled Eager AutoCodeGen for operators with multiple OpBases
      
      * Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument
      
      * Handled Dispensable Inputs/Outputs in Eager AutoCodeGen
      
      * Adjusted function generation/call between Python-C API & Dygraph API
      
      * Synchronized auto-generated Python-C API with Dygraph Forward Functions
      
      * support more eager tensor api
      
      * fix merge compile error
      
      * fix compile error and fit develop code
      
      * support pure CPU
      
      * fix some logic error in eager_mode
      
      * support _varbase_creator in eager mode
      
      * Added safe_initialized interface to EagerTensor for use in processing dispensable inputs
      
      * for eager mode
      
      * refine
      
      * support multiple constructor for eager tensor
      
      * add place related code
      
      * polish code
      
      * specific randint with dtype of int64
      
      * Support pure cpu test
      
      * eager logic
      
      * refine test in pure cpu
      
      * eager logic
      
      * eager logic
      
      * eager logic, test=develop
      
      * skip core.eager when in inference, test=develop
      
      * refine, test=develop
      
      * refine, test=develop
      
      * call RetainGrad after run forward kernel, test=develop
      
      * refine, test=develop
      
      * support dygraph util, meta, guard test
      
      * support inference test
      
      * refine test and fix initializer failed
      Co-authored-by: Njim19930609 <jim19930609@gmail.com>
      Co-authored-by: NWang Huan <wanghuan29@baidu.com>
      d48f7c89
  4. 02 12月, 2021 1 次提交
  5. 22 3月, 2021 1 次提交
  6. 26 1月, 2021 1 次提交
  7. 01 12月, 2020 1 次提交
  8. 03 9月, 2020 1 次提交
  9. 15 8月, 2020 1 次提交
    • Z
      expose and unify the Tensor concepts to the user (#25978) · 6de463d3
      Zhou Wei 提交于
      * expose and unify the Tensor concepts to the user
      
      * expose tensor to user
      
      * add copy place for Tensor
      
      * add copy place for Tensor
      
      * add note
      
      * add macro PADDLE_WITH_CUDA
      
      * remove RUN_TYPE=DIST
      
      * fix some error
      6de463d3
  10. 30 7月, 2020 1 次提交
  11. 19 6月, 2020 1 次提交
  12. 14 5月, 2020 1 次提交
  13. 08 5月, 2020 1 次提交
  14. 10 3月, 2020 1 次提交
  15. 23 2月, 2020 1 次提交
  16. 06 2月, 2020 1 次提交
  17. 22 1月, 2020 1 次提交
  18. 16 12月, 2019 1 次提交
  19. 10 12月, 2019 1 次提交
  20. 14 11月, 2019 1 次提交
  21. 24 10月, 2019 1 次提交
  22. 14 10月, 2019 1 次提交
    • Z
      add input type and dtype check for cast_op (#20070) · 01209b51
      zhupengyang 提交于
      * add input type and dtype check for cast_op
      test=develop
      
      * fix annotation
      test=develop
      
      * support more data type
      
      test=develop
      
      * fix bug for fill_constant's error type
      
      test=develop
      
      * improve converage
      
      test=develop
      
      * improve converage
      
      test=develop
      01209b51
  23. 12 10月, 2019 1 次提交
  24. 11 10月, 2019 1 次提交
  25. 03 10月, 2019 1 次提交
    • W
      Fill constant error message fix (#20075) · 9a76f3f9
      wangchaochaohu 提交于
      * fix the constant error message test=develop
      
      * fix typo test=develop
      
      * fix typo test=develop
      
      * fix code style test=develop
      
      * fix comment and bugs test=develop
      
      * fix the bug test=develop
      
      * fix and add unittest test=develop
      
      * fix the typo test=develop
      
      * add support for the fill_constant op test=develop
      
      * add test for ci coverage test=develop
      9a76f3f9
  26. 03 9月, 2019 1 次提交
  27. 16 8月, 2019 1 次提交
  28. 21 6月, 2019 1 次提交
  29. 13 6月, 2019 1 次提交
  30. 06 6月, 2019 1 次提交
  31. 16 5月, 2019 1 次提交
  32. 19 3月, 2019 2 次提交
  33. 01 3月, 2019 1 次提交
  34. 23 1月, 2019 1 次提交
  35. 09 1月, 2019 1 次提交
  36. 07 1月, 2019 1 次提交
  37. 27 12月, 2018 1 次提交
  38. 26 12月, 2018 1 次提交
    • W
      Fp16 training (#14992) · 856f0da0
      Wu Yi 提交于
      * wip
      
      * wip
      
      * wip
      
      * wip for test
      
      * add fp16 tests test=develop
      
      * fix cpu build test=develop
      
      * fix test=develop
      
      * fix py3 tests test=develop
      
      * fix lr_scheduler dtype test=develop
      
      * fix test=dvelop
      
      * test fix ci compile test=develop
      
      * fix build and merge test=develop
      
      * fallback momentumop change to general test=develop
      
      * make fp16 lr schedule simple test=develop
      
      * fix ut test=develop
      
      * fix tests test=develop
      
      * remove fp16 learning rate cast test=develop
      856f0da0
  39. 20 12月, 2018 1 次提交