1. 10 1月, 2019 3 次提交
    • X
      Conv int8 residual (#15145) · 8f17c714
      xiaolil1 提交于
      * Enable basic MKL-DNN INT8 Conv OP
      test=develop
      
      * Modify test case
      test=develop
      
      * Clean unittest code
      test=develop
      
      * Fix test
      test=develop
      
      * Modify test
      test=develop
      
      * Enable MKL-DNN INT8 Conv with Relu Fusion OP
      test=develop
      
      * Enable INT8 Conv with residual fusion OP
      test=develop
      
      * Modify code.
      test=develop
      
      * Modify basic INT8 Conv
      test=develop
      
      * Modify Conv.
      test=develop
      
      * fix style
      test=develop
      
      * Fix style
      test=develop
      
      * Fix test
      test=develop
      
      * Modify code.
      test=develop
      
      * Fix test
      test=develop
      8f17c714
    • W
      [Feature] support mix precision training for resnet (#14899) · fd854183
      Wu Yi 提交于
      * clip softmax for fp16
      
      * updates
      
      * fuse xent support fp16 test=develop
      
      * wip
      
      * wip
      
      * add simple row reduce
      
      * wip fp16 accurate softmax
      
      * add accurate softmax kernel for fp16 test=develop
      
      * update test=develop
      
      * fix cpu build test=develop
      
      * update api.spec test=develop
      
      * follow comments test=develop
      
      * fix build test=develop
      
      * fix trt build test=develop
      
      * fix inference build test=develop
      
      * fix merge test=develop
      
      * update test=develop
      
      * try fix build test=develop
      
      * fix build test=develop
      
      * rename real_exp test=develop
      
      * fortest
      
      * remove hacky kernels test=develop
      
      * clean up test=develop
      fd854183
    • T
      follow comment and fix typo · 8e086a85
      tensor-tang 提交于
      test=develop
      8e086a85
  2. 09 1月, 2019 1 次提交
  3. 08 1月, 2019 5 次提交
  4. 07 1月, 2019 16 次提交
  5. 04 1月, 2019 6 次提交
  6. 03 1月, 2019 1 次提交
  7. 02 1月, 2019 3 次提交
  8. 29 12月, 2018 5 次提交