1. 10 12月, 2019 1 次提交
    • W
      modify static_kernel_pass to support select the kernel according to input type (#2488) · 7ef0e7fe
      Wilber 提交于
      修改了选kernel的逻辑,默认从模型文件中读取出lod_tensor的data type,在static_kernel_pick pass中如果kernel输入输出的类型与读取的data type完全一致,则选择该Kernel的概率增大。
      
      - 增加 从模型文件__model__读取lod_tensor的data type到cpp::vardesc
      
      - program中增加unordered_map<string, type>字段,并在 Program::PrepareWorkspace中对该字段赋值
      
      - 修改了node.h文件,将const Type* 更改为Type*,并在SSAGraph::Build过程中为符合条件的type*赋值
      
      - static_kernel_pick_pass中添加新规则,如果kernel的输入类型输出类型与__model__中存储的类型的一致,则score*=2。
      
      - 支持模型中用到sequence_reverse_float kernel(输入输出均为float)和sequence_reverse_int64 kernel(输入输出均为int64),能够根据输入输出type选kernel
      7ef0e7fe
  2. 07 12月, 2019 1 次提交
    • J
      Support mask_rcnn (#2484) · c2f72cb3
      juncaipeng 提交于
      * add arm split lod tensor, test=develop
      
      * add arm merge lod tensor, test=develop
      
      * update split merge lod tensor, test=develop
      
      * add reduce_prob op, test=develop
      
      * support mask_rcnn succeed, test=develop
      c2f72cb3
  3. 04 12月, 2019 1 次提交
  4. 30 10月, 2019 1 次提交
  5. 24 10月, 2019 1 次提交
    • L
      Make inceptionv4, resnet50, googlenet can run on x86 paltform (#2250) · edb4ea9a
      liu zhengxi 提交于
      * make inceptionv4, resnet50, googlenet can run on x86 paltform and fix the compare part in x86 unittests, test=develop
      
      * fix googlenet tests for benchmark record, test=develop
      
      * [framework][profile] fix profile dump bug when op is feed and fetch test=develop (sangoly)
      edb4ea9a
  6. 16 10月, 2019 1 次提交
  7. 27 9月, 2019 2 次提交
  8. 19 9月, 2019 1 次提交
  9. 01 9月, 2019 1 次提交
    • Y
      [ARM][CPU] Fix time counter of arm cpu profiler (#1925) · e3fb95ae
      Yuan Shuai 提交于
      * Fix timer of arm cpu profiler. test=develop
      
      * Fix un-added op in cmake.test=develop
      
      * fix cmake error
      
      * fix cmake error, test=develop
      
      * Fix pass sequence. test=develop
      
      * replace option with lite_option. test=develop
      
      * disable profile mode by default. test=develop
      
      * Fix error option name. test=develop
      e3fb95ae
  10. 30 8月, 2019 1 次提交
    • Z
      add precision and persistable attrs for the tensor. (#1899) · e2e07fa4
      Zhen Wang 提交于
      * Add precision and persistable attrs for the tensor. And fix cxx light and full api demo.
      
      * update precision2string methods. test=develop
      
      * move the save logic to the front of the run in mobilenetv1_full_api.cc, test=develop.
      
      * add comments for UpdateVarsOfProgram. test=develop
      e2e07fa4
  11. 16 8月, 2019 1 次提交