1. 12 4月, 2019 1 次提交
  2. 11 4月, 2019 1 次提交
  3. 30 3月, 2019 2 次提交
  4. 29 3月, 2019 16 次提交
  5. 28 3月, 2019 4 次提交
  6. 27 3月, 2019 1 次提交
  7. 26 3月, 2019 4 次提交
  8. 25 3月, 2019 2 次提交
  9. 24 3月, 2019 1 次提交
  10. 21 3月, 2019 1 次提交
    • S
      add more unittest · 953214ad
      sneaxiy 提交于
      modify allocator strategy
      remove changes of legacy buddy_allocator
      test=develop
      953214ad
  11. 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
  12. 19 3月, 2019 2 次提交
  13. 18 3月, 2019 2 次提交
  14. 15 3月, 2019 2 次提交
    • M
      Implement Runtime Var Type Inference · 438bca9c
      minqiyang 提交于
      test=develop
      438bca9c
    • 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