1. 02 4月, 2020 1 次提交
  2. 07 1月, 2020 1 次提交
  3. 02 1月, 2020 1 次提交
  4. 01 12月, 2019 1 次提交
  5. 28 11月, 2019 1 次提交
    • K
      batch_norm momentum support variable (#21246) · 67c836fb
      Kaipeng Deng 提交于
      * batch_norm momentum support variable. test=develop
      
      * fix format. test=develop
      
      * add batch_norm momentum variable example. test=develop
      
      * move MomentumTensor to training branch. test=develop
      
      * split example. test=develop
      
      * fix doc. test=develop
      
      * fix PADDLE_ENFORCE ci. test=develop
      
      * fix format. test=develop
      67c836fb
  6. 26 11月, 2019 1 次提交
  7. 25 9月, 2019 1 次提交
  8. 19 8月, 2019 1 次提交
  9. 16 8月, 2019 1 次提交
  10. 19 6月, 2019 1 次提交
    • fix spelling errors (#17941) · 802ea509
      翟飞跃 提交于
      * fix spelling errors; test=develop
      
      * Update API.spec
      
      update md5
      
      * Update API.spec
      
      * change the order of api;test=develop
      802ea509
  11. 31 3月, 2019 1 次提交
    • W
      Batch norm cudnn accurate (#16545) · 22b02bfa
      Wu Yi 提交于
      * fix cudnn batch norm accuracy test=develop
      
      * fix cudnn batch norm accuracy test=develop
      
      * disable failed test for later fix test=develop
      22b02bfa
  12. 21 3月, 2019 1 次提交
  13. 20 3月, 2019 1 次提交
  14. 18 3月, 2019 1 次提交
  15. 15 3月, 2019 1 次提交
    • Q
      Support sync batch norm. (#16121) · 8ad672a2
      qingqing01 提交于
      * Support Sync Batch Norm.
      * Note, do not enable it in one device.
      
      Usage:
      
      build_strategy = fluid.BuildStrategy()
      build_strategy.sync_batch_norm = True
      binary = fluid.compiler.CompiledProgram(tp).with_data_parallel(
              loss_name=loss_mean.name,
              build_strategy=build_strategy)
      8ad672a2
  16. 29 11月, 2018 1 次提交
  17. 26 11月, 2018 1 次提交
  18. 08 11月, 2018 1 次提交
  19. 07 11月, 2018 1 次提交
    • C
      Add fp16 backward support (#14202) · a9b5d42d
      chengduo 提交于
      * add fp16 backward support
      test=develop
      
      * add sum_op fp16 test
      
      * disable test_dist_save_load
      test=develop
      
      * add check_grad for sum
      
      * add unit test for softmax_grad fp16
      test=develop
      
      * add scale_op unit test
      
      * add mul_grad_op unit test for fp16
      
      * add cross_entropy_grad and eman_grad unit test for fp16
      test=develop
      
      * fix cross_entropy unit test
      
      * add pool2d fp16 unit test
      
      * refine conv2d fp16 unit test
      test=develop
      
      * refine activation unit test
      test=develop
      
      * fix ci
      test=develop
      
      * follow zhihong's comment, copy from https://github.com/PaddlePaddle/Paddle/pull/12796
      test=develop
      a9b5d42d
  20. 05 11月, 2018 1 次提交
  21. 24 10月, 2018 1 次提交
  22. 10 7月, 2018 1 次提交
  23. 02 5月, 2018 1 次提交
  24. 12 4月, 2018 1 次提交
  25. 11 4月, 2018 1 次提交
  26. 19 3月, 2018 3 次提交
  27. 18 3月, 2018 1 次提交
  28. 17 3月, 2018 1 次提交
  29. 12 2月, 2018 1 次提交
  30. 10 2月, 2018 2 次提交
  31. 25 12月, 2017 1 次提交
  32. 22 12月, 2017 1 次提交
  33. 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
  34. 28 11月, 2017 1 次提交
  35. 11 11月, 2017 1 次提交
  36. 28 10月, 2017 2 次提交