1. 29 10月, 2018 1 次提交
    • W
      [1.1] [project] train imagenet using large batch size (#13766) · 26200f2e
      Wu Yi 提交于
      * fix nccl2 lars dist support
      
      * put lars in momentum op
      
      * add tests lars
      
      * fix ci
      
      * fix cpu kernel
      
      * soft warning
      
      * remove lars in test_recognize_digits.py
      
      * move to another op
      
      * add file
      
      * update api.spec test=develop
      
      * update test=develop
      
      * fix api.spec test=develop
      
      * wip
      
      * wip, finish grad merge ops
      
      * wip, finish graph build
      
      * wip test running
      
      * work on 1 gpu
      
      * workable version
      
      * update
      
      * fix tests
      
      * fuse broadcast op
      
      * fix compile failed
      
      * refine
      
      * add batch merge test mnist
      
      * fix CI test=develop
      
      * fix build
      
      * use independent bn params for batch merge test=develop
      
      * update api.spec
      
      * follow comments and for test
      
      * wip
      
      * refine tests test=develop
      
      * follow comments test=develop
      
      * remove startup bn modify test=develop
      
      * follow comments test=develop
      
      * fix merge test=develop
      26200f2e
  2. 12 9月, 2018 1 次提交
  3. 31 8月, 2018 1 次提交
  4. 29 8月, 2018 1 次提交
  5. 21 8月, 2018 1 次提交
  6. 09 8月, 2018 1 次提交
  7. 26 7月, 2018 4 次提交
  8. 22 7月, 2018 1 次提交
  9. 19 7月, 2018 1 次提交
  10. 18 7月, 2018 4 次提交
  11. 26 6月, 2018 1 次提交
  12. 11 6月, 2018 1 次提交
  13. 08 6月, 2018 1 次提交
  14. 06 6月, 2018 1 次提交
    • Y
      SSA Graph Builder Factory · d9af1532
      yuyang18 提交于
      * Use Builder Chain to decorate new builders. It is easy to extend
        builders.
      * Make graphviz path as a build strategy, not a FLAGS.
      d9af1532
  15. 15 5月, 2018 1 次提交
  16. 14 5月, 2018 2 次提交
  17. 07 4月, 2018 1 次提交
  18. 07 3月, 2018 2 次提交
  19. 06 3月, 2018 2 次提交
  20. 15 2月, 2018 1 次提交
    • Y
      Update tensor_util.h (#8422) · cfffb1a3
      Yi Wang 提交于
      * Update tensor_util.h
      
      * Update with moved TensorDesc
      
      * Fix tensur_utils.cu
      
      * Update
      
      * Update
      
      * Update
      
      * Update
      
      * Make tensor_util.cu a symbolic link
      cfffb1a3
  21. 10 2月, 2018 2 次提交
  22. 07 2月, 2018 1 次提交
  23. 06 2月, 2018 2 次提交
  24. 01 2月, 2018 1 次提交
  25. 31 1月, 2018 1 次提交
  26. 30 1月, 2018 1 次提交