1. 21 4月, 2019 1 次提交
    • Z
      Refine model gpu memory (#16993) · 1202d3fc
      Zeng Jinle 提交于
      * speedup gc and inplace softmax_with_cross_entropy_grad
      test=develop
      
      * refine models gpu mem
      Merge skip vars and warning messages of mem opt
      remove relu mem opt
      test=develop
      
      * follow comments
      test=develop
      1202d3fc
  2. 12 4月, 2019 1 次提交
  3. 11 4月, 2019 1 次提交
  4. 08 4月, 2019 2 次提交
  5. 03 4月, 2019 1 次提交
  6. 28 3月, 2019 2 次提交
  7. 22 3月, 2019 1 次提交
    • C
      [Speed]Refine ParallelExecutor (#16190) · a6a3b2fb
      chengduo 提交于
      * refine parallelExecutor
      test=develop
      
      * Polish op_handle
      test=develop
      
      * Remove unnecessary op_handle
      test=develop
      
      * Fix Travis CI
      test=develop
      
      * Fix fetch bug
      test=develop
      
      * Remove WaitInputVarGenerated
      
      * Fix OpHandleBase::Run
      test=develop
      
      * debug
      test=develop
      
      * use origin fetch_op_handle
      test=develop
      
      * Revert op_handle_base.cc
      test=develop
      
      * Polish code
      test=develop
      
      * Fix OpHandleBase::Run
      test=develop
      
      * code refine
      
      * test CI and CE
      test=develop
      
      * fix OpHandle::Run
      test=develop
      
      * refine AllReduceOpHandle
      test=develop
      
      * Polish code
      test=develop
      a6a3b2fb
  8. 20 3月, 2019 1 次提交
    • C
      Fuse AllReduce (#15921) · f26ba5bd
      chengduo 提交于
      * fuse all_reduce
      test=develop
      
      * add fuse_parameter_groups_size
      test=develop
      
      * Polish code
      test=develop
      
      * Fix travis-ci
      test=develop
      
      * Add SetGroupAccordingToLayers and SetGroupAccordingToGroupSize
      test=develop
      
      * Add SetGroupAccordingToMemorySize
      test=develop
      
      * fix multi_devices_graph
      test=develop
      
      * reset params_grads
      test=develop
      
      * Polish code
      test=develop
      f26ba5bd
  9. 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
  10. 07 3月, 2019 1 次提交
  11. 05 3月, 2019 1 次提交
  12. 23 2月, 2019 1 次提交
  13. 22 2月, 2019 2 次提交
  14. 21 2月, 2019 2 次提交
  15. 19 2月, 2019 1 次提交
  16. 18 2月, 2019 2 次提交
  17. 14 2月, 2019 3 次提交
  18. 12 2月, 2019 2 次提交
  19. 11 2月, 2019 2 次提交
  20. 08 2月, 2019 1 次提交
  21. 07 2月, 2019 1 次提交
  22. 31 1月, 2019 2 次提交
  23. 30 1月, 2019 1 次提交
  24. 27 1月, 2019 1 次提交
  25. 21 1月, 2019 2 次提交
    • D
      squash commits. test=develop · 8f3b2523
      dzhwinter 提交于
      8f3b2523
    • D
      Memory optimization of depthwise conv op and group norm op (#15313) · 9f8f0fc2
      Dun 提交于
      * mem opt
      
      * test=develop
      
      * test=develop
      
      * test=develop
      
      * test=develop
      
      * test=develop
      
      * test=develop
      
      * test=develop
      
      * refine code  test=develop
      
      * refine code  test=develop
      
      * refine code  test=develop
      
      * refine code  test=develop
      
      * refine with cub test=develop
      
      * fix mkldnn test && remove comments && test=develop
      
      * polish code && test=develop
      
      * add only_forward test && test=develop
      9f8f0fc2
  26. 20 1月, 2019 1 次提交
  27. 19 1月, 2019 1 次提交
  28. 17 1月, 2019 1 次提交
  29. 07 1月, 2019 1 次提交