1. 29 1月, 2019 1 次提交
  2. 21 1月, 2019 1 次提交
    • 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
  3. 14 1月, 2019 1 次提交
  4. 13 1月, 2019 1 次提交
  5. 11 1月, 2019 1 次提交
  6. 10 1月, 2019 1 次提交
  7. 08 1月, 2019 1 次提交
  8. 07 1月, 2019 1 次提交
  9. 25 12月, 2018 1 次提交
  10. 16 12月, 2018 1 次提交
  11. 14 12月, 2018 1 次提交
  12. 07 12月, 2018 1 次提交
  13. 03 12月, 2018 1 次提交
  14. 15 11月, 2018 1 次提交
    • S
      add mkldnn prop_kind phase for inference-only case to pooling and activations (#14278) · 8a1eeec5
      Sylwester Fraczek 提交于
      * add is_test to pooling and activations
      
      add prop_kind support for layers activation. conv and pooling
      
      add a pass that sets is_test to true
      
      add transpiler version of is_test pass
      
      test=develop
      
      * patch test and pass
      
      test=develop
      
      * add pass to analyzer.h
      
      test=develop
      
      * add is_test attr description & pass only on mkldnn
      
      in:
      activation_op.cc
      batch_norm_op.cc
      conv_op.cc
      dropout_op.cc
      lrn_op.cc
      pool_op.cc
      sequence_pool_op.cc
      softmax_op.cc
      
      * fix is_test handling for activation pool and conv
      
      * change description of is_test for all layers again
      
      * remove GetAttr(use_mkldnn) from pass
      
      * rename correct_mkldnn_test_phase to is_test
      
      and remove dependency on MKLDNN
      test=develop
      
      * review fix magic number
      
      * two if(..)s into one
      
      * Check is_test once and pass mkldnn forward prop kind
      
      * dereference shared_ptr with * (without get())
      
      test=develop
      
      * add is_test_pass back
      
      test=develop
      8a1eeec5
  15. 14 11月, 2018 1 次提交
  16. 06 11月, 2018 1 次提交
  17. 31 10月, 2018 1 次提交
    • S
      add depthwise conv mkldnn pass · 4e2aaf01
      Sylwester Fraczek 提交于
      added depthwise conv mkldnn pass which for MKLDNN changes depthwise_conv operator to conv operator because for mkldnn this is the same api
      test=develop
      4e2aaf01
  18. 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
  19. 23 10月, 2018 1 次提交
  20. 21 10月, 2018 2 次提交
  21. 19 10月, 2018 3 次提交
  22. 11 10月, 2018 1 次提交
  23. 10 10月, 2018 1 次提交
  24. 08 10月, 2018 1 次提交
    • S
      conv bn fuse pass · 78f98294
      Sylwester Fraczek 提交于
      review fix
      
      review from hshen14 fix
      
      test=develop
      
      fix error in broadcast and code cleanup
      
      rename bias -> eltwise and added macro to shorten code
      
      formatting
      78f98294
  25. 29 9月, 2018 1 次提交
  26. 28 9月, 2018 1 次提交
  27. 27 9月, 2018 1 次提交
    • J
      - Added initial pass for embedding-fc-lstm · 7ab5626d
      Jacek Czaja 提交于
      - Added draft of new operator
      
      - Added fused embedding fc lstm files
      
      - First time embedding_fc_lstm_fuse_pass was invoked in
        test_text_classification
      
      - Added Embedding pattern
      
      - Not crashing
      
      - Enabled draft of embedding_fc_lstm pass (does it job)
      
      - First working (Seqcompute only) version
      
      - Removed diagnostic comment
      
      - First enabling of BatchCompute
      
      - Disabling pass for embedding with is_sparse and is_distributed
      
      - Cosmetics
      
      - Style
      
      - Style
      7ab5626d
  28. 25 9月, 2018 1 次提交
  29. 20 9月, 2018 1 次提交
    • C
      Feature/op_fuse_pass (#12440) · d402234b
      chengduo 提交于
      * Add Preface
      
      * Add demo code
      
      * Save file
      
      * Refine code
      
      * seems can work
      
      * use elementwise strategy
      
      * Use ElementwiseComputeEx
      
      * Add comments
      
      * extract functions from operator
      
      * Refine code
      
      * Follow comment
      
      * code refine
      
      * add op_fuse  pass
      
      * add backward
      
      * code refine
      
      * use TopologySortOperations
      
      * follow comments
      
      * refine IsFusible
      
      * code enhance
      
      * fix op_fusion_pass
      
      * refine code
      
      * refine fuse_elemwise_act_op
      
      * adjust the input and output
      
      * refine logic
      
      * add intermediate_edge
      
      * disable inplace
      
      * follow comments
      
      * refine logic
      
      * follow comments
      
      * Remove the removable IntermediateOut
      
      * change strategy
      
      * code refine
      
      * enable fuse backward
      
      * code refine
      
      * code refine
      
      * rename unit test
      
      * follow comments
      d402234b
  30. 12 9月, 2018 1 次提交
  31. 10 9月, 2018 1 次提交
  32. 07 9月, 2018 1 次提交
  33. 06 9月, 2018 1 次提交
  34. 03 9月, 2018 2 次提交
  35. 01 9月, 2018 1 次提交
  36. 29 8月, 2018 1 次提交