1. 02 4月, 2019 1 次提交
  2. 30 3月, 2019 1 次提交
  3. 28 3月, 2019 2 次提交
  4. 25 3月, 2019 1 次提交
  5. 22 3月, 2019 1 次提交
  6. 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
  7. 09 3月, 2019 1 次提交
    • B
      Upgrade MKLDNN to v0.18-rc and fix issue caused by lib/lib64 (#15861) · db120b93
      Brian Liu 提交于
      * Upgrade MKLDNN to v0.18-rc and fix issue caused by lib/lib64
      
      Upgrade MKLDNN to v0.18-rc
      Also fix the issue during upgrade
      
      test=develop
      
      * Rebase MKLDNN to rls-v0.18 branch
      
      Some issues in v0.18-rc which caused INT8 conv op unit test failure was fixed
      in rls-v0.18 branch
      
      test=develop
      
      * Upgrade MKLDNN from v0.18rc to formal v0.18 tag
      
      test=develop
      
      * Fix the windows compile issue.
      
      test=develop
      db120b93
  8. 04 3月, 2019 4 次提交
  9. 28 2月, 2019 1 次提交
  10. 27 2月, 2019 1 次提交
    • D
      polish cudnn related code and fix bug. (#15164) · 225c11a9
      dzhwinter 提交于
      * staged.
      
      * polish code
      
      * polish code. test=develop
      
      * polish code. test=develop
      
      * api change. test=develop
      
      * fix default value. test=develop
      
      * fix default value. test=develop
      225c11a9
  11. 26 2月, 2019 2 次提交
  12. 25 2月, 2019 1 次提交
  13. 22 2月, 2019 2 次提交
    • T
      Revert 15770 develop a6910f90 gelu mkl opt (#15872) · ee2321de
      tensor-tang 提交于
      * Revert "Optimze Gelu with MKL Erf function (#15770)"
      
      This reverts commit 676995c8.
      
      * test=develop
      ee2321de
    • Y
      Optimze Gelu with MKL Erf function (#15770) · 676995c8
      Yihua Xu 提交于
      * Optimize for gelu operator
      
      * Set up the low accuracy mode of MKL ERF function.
      
      test=develop
      
      * Only enable MKLML ERF when OS is linux
      
      * Use the speical mklml version included vmsErf function to verify gelu mkl kernel.
      
      test=develop
      
      * Add the CUDA macro to avoid NVCC's compile issue.
      
      test=develop
      
      * Add the TODO comments for mklml library modification.
      
      test=develop
      
      * Clean Code
      
      test=develop
      
      * Add the comment of marco for NVCC compiler.
      
      test=develop
      676995c8
  14. 20 2月, 2019 3 次提交
  15. 19 2月, 2019 7 次提交
  16. 15 2月, 2019 1 次提交
  17. 14 2月, 2019 1 次提交
  18. 12 2月, 2019 1 次提交
  19. 11 2月, 2019 1 次提交
  20. 01 2月, 2019 3 次提交
  21. 29 1月, 2019 1 次提交
  22. 25 1月, 2019 2 次提交
  23. 21 1月, 2019 1 次提交