提交 · d3d793c71fe60c2bd1a0ff8e5d1e4ca4ef12e1ae · PaddlePaddle / Paddle-Lite

22 5月, 2020 2 次提交
- H
  
  [cherry-pick][BUG FIX] fix the issue that opt can not convert quantized model (#3683) · 6dcbe32b
  由 huzhiqiang 提交于 5月 22, 2020
  
  6dcbe32b
- H
  
  [cherry-pick][BUG FIX] fix the issue that opt can not convert quantized model (#3683) · f779a5b9
  由 huzhiqiang 提交于 5月 22, 2020
  
  f779a5b9
21 5月, 2020 1 次提交
- C
  
  mereged with upstream/develope · e5f6ce88
  由 chonwhite 提交于 5月 21, 2020
  
  e5f6ce88
18 5月, 2020 4 次提交
- Y
  [LITE][OPENCL] Enhance Profiler for OpenCL with in/out/filter shape,... · e18ba102
  由 Yuan Shuai 提交于 5月 18, 2020
```
[LITE][OPENCL] Enhance Profiler for OpenCL with in/out/filter shape, macs/macs_ps, real backend kernel etc. (#3641)

* [LITE][OPENCL] Enhance Precision Profiler for OpenCL. test=develop
```
  e18ba102
- Y
  [LITE][OPENCL] Enhance Profiler for OpenCL with in/out/filter shape,... · 53a6f3bc
  由 Yuan Shuai 提交于 5月 18, 2020
```
[LITE][OPENCL] Enhance Profiler for OpenCL with in/out/filter shape, macs/macs_ps, real backend kernel etc. (#3641)

* [LITE][OPENCL] Enhance Precision Profiler for OpenCL. test=develop
```
  53a6f3bc
- H
  
  [Framework][ModelType] Add Shape&Precision information into optimized model (#3643) · 05690463
  由 huzhiqiang 提交于 5月 18, 2020
  
  05690463
- H
  
  [Framework][ModelType] Add Shape&Precision information into optimized model (#3643) · 546d4da8
  由 huzhiqiang 提交于 5月 18, 2020
  
  546d4da8
15 4月, 2020 2 次提交
- M
  refactor(*): reduce Wsign-compare warning (#3391) · 6fe95238
  由 MaxwellDing 提交于 4月 15, 2020
```
refactor(*): reduce Wsign-compare warning
```
  6fe95238
- M
  refactor(*): reduce Wsign-compare warning (#3391) · 2997b937
  由 MaxwellDing 提交于 4月 15, 2020
```
refactor(*): reduce Wsign-compare warning
```
  2997b937
13 4月, 2020 2 次提交
- W
  lite cuda support exec multi-stream. (#2949) · 5382994e
  由 Wilber 提交于 4月 13, 2020
```
lite cuda support exec multi-stream
```
  5382994e
- W
  lite cuda support exec multi-stream. (#2949) · 4a7284f9
  由 Wilber 提交于 4月 13, 2020
```
lite cuda support exec multi-stream
```
  4a7284f9
03 4月, 2020 2 次提交
- Y
  [LITE][PROFILER] Split precision profiler from performance profiler (#3305) · 6b53c0ca
  由 Yuan Shuai 提交于 4月 03, 2020
```
* split precision profiler from performance profiler. test=develop
```
  6b53c0ca
- Y
  [LITE][PROFILER] Split precision profiler from performance profiler (#3305) · 185c7096
  由 Yuan Shuai 提交于 4月 03, 2020
```
* split precision profiler from performance profiler. test=develop
```
  185c7096
31 3月, 2020 2 次提交
- H
  
  [operator] add InferShapeImpl method (#3294) · c754a38f
  由 huzhiqiang 提交于 3月 31, 2020
  
  c754a38f
- H
  
  [operator] add InferShapeImpl method (#3294) · 1f8b5c2b
  由 huzhiqiang 提交于 3月 31, 2020
  
  1f8b5c2b
25 3月, 2020 2 次提交
- X
  fix: fix infershape profile (#3240) · 50152b80
  由 xiaogang 提交于 3月 25, 2020
```
test=develop
```
  50152b80
- X
  fix: fix infershape profile (#3240) · 6a0a1f08
  由 xiaogang 提交于 3月 25, 2020
```
test=develop
```
  6a0a1f08
22 3月, 2020 2 次提交

[LITE][OPENCL][PROFILE] Enhance precision profile & Clean opencl code (#3227) · 5b2a618f

由 Yuan Shuai 提交于 3月 22, 2020

* [LITE][OPENCL] clean code for opencl. test=develop

* [LITE][PROFILER] Enhance Precision Profiler. test=develop

* delete useless var in profiler. test=develop

* add ocl header. test=develop

5b2a618f

[LITE][OPENCL][PROFILE] Enhance precision profile & Clean opencl code (#3227) · 3868be2c

由 Yuan Shuai 提交于 3月 22, 2020

* [LITE][OPENCL] clean code for opencl. test=develop

* [LITE][PROFILER] Enhance Precision Profiler. test=develop

* delete useless var in profiler. test=develop

* add ocl header. test=develop

3868be2c

17 3月, 2020 2 次提交

add cuda cxx demo (#3205) · 9098da7c

由 Wilber 提交于 3月 17, 2020

- 增加cuda c++ demo.
- 考虑到检测模型尾部一般是multiclass_nms，该kernel为host，如果fetch kernel为cuda的话，则会在此处插入无用的io_copy(host->cuda)，由于该原因，注释掉fetch的cuda kernel. 默认使用host的fetch kernel. 此处暗中进行的行为：每次predictor run完，都会默认把数据从cuda拷贝到cpu

9098da7c

add cuda cxx demo (#3205) · f6461e39

由 Wilber 提交于 3月 17, 2020

- 增加cuda c++ demo.
- 考虑到检测模型尾部一般是multiclass_nms，该kernel为host，如果fetch kernel为cuda的话，则会在此处插入无用的io_copy(host->cuda)，由于该原因，注释掉fetch的cuda kernel. 默认使用host的fetch kernel. 此处暗中进行的行为：每次predictor run完，都会默认把数据从cuda拷贝到cpu

f6461e39

20 2月, 2020 2 次提交
- W
  Optimize cuda kernel and remove io_copy added by default due to missing fetch_cuda kernel (#2920) · 6e76d91e
  由 Wilber 提交于 2月 20, 2020
```
Optimize cuda kernel and remove io_copy added by default due to missing fetch_cuda kernel
```
  6e76d91e
- W
  Optimize cuda kernel and remove io_copy added by default due to missing fetch_cuda kernel (#2920) · 823f0dae
  由 Wilber 提交于 2月 20, 2020
```
Optimize cuda kernel and remove io_copy added by default due to missing fetch_cuda kernel
```
  823f0dae
17 2月, 2020 1 次提交
- T
  
  delete unused code test=develop · 10d6e488
  由 tienfeek 提交于 2月 17, 2020
  
  10d6e488
14 2月, 2020 3 次提交
- T
  FPGA support quantitative model · ffb7ead1
  由 tienfeek 提交于 2月 14, 2020
```
test=develop
```
  ffb7ead1
- X
  fix: fix fpga run the feed/fetch op (#2868) · a95a389c
  由 xiaogang 提交于 2月 14, 2020
```
fix fpga lite_tensor compile bug
     add fake quantize_abs_max op
     test=develop
```
  a95a389c
- X
  fix: fix fpga run the feed/fetch op (#2868) · 27ec5deb
  由 xiaogang 提交于 2月 14, 2020
```
fix fpga lite_tensor compile bug
     add fake quantize_abs_max op
     test=develop
```
  27ec5deb
30 12月, 2019 2 次提交
- Y
  Optimize the execution of RuntimeProgram by saving the bool whether the op is... · 383ed8f5
  由 Yiqun Liu 提交于 12月 30, 2019
```
Optimize the execution of RuntimeProgram by saving the bool whether the op is feed/fetch op. (#2703)

test=develop
```
  383ed8f5
- Y
  Optimize the execution of RuntimeProgram by saving the bool whether the op is... · bb1cf7ff
  由 Yiqun Liu 提交于 12月 30, 2019
```
Optimize the execution of RuntimeProgram by saving the bool whether the op is feed/fetch op. (#2703)

test=develop
```
  bb1cf7ff
27 12月, 2019 3 次提交
- C
  
  merged from 1.4.0 · 64a0feac
  由 chonwhite 提交于 12月 27, 2019
  
  64a0feac
- 石
  
  update profiler, test=develop (#2644) · e9bdfb67
  由石晓伟提交于 12月 27, 2019
  
  e9bdfb67
- 石
  
  update profiler, test=develop (#2644) · 9171b70e
  由石晓伟提交于 12月 27, 2019
  
  9171b70e
26 12月, 2019 1 次提交
- M
  
  fix:delete useless code · 32177924
  由 MyPandaShaoxiang 提交于 12月 26, 2019
  
  32177924
23 12月, 2019 2 次提交
- H
  
  [lite][arm]fix model_optimize bug, update concat and split op, speed up (#2620) · 186397fe
  由 HappyAngel 提交于 12月 23, 2019
  
  186397fe
- H
  
  [lite][arm]fix model_optimize bug, update concat and split op, speed up (#2620) · 6946ca23
  由 HappyAngel 提交于 12月 23, 2019
  
  6946ca23
19 12月, 2019 2 次提交
- Y
  [ARM] change global pooling choose kernel policy, test=develop (#2602) · 187bf7c5
  由 yiicy 提交于 12月 19, 2019
```
* [ARM] change global pooling choose kernel policy, test=develop
```
  187bf7c5
- Y
  [ARM] change global pooling choose kernel policy, test=develop (#2602) · 49f03648
  由 yiicy 提交于 12月 19, 2019
```
* [ARM] change global pooling choose kernel policy, test=develop
```
  49f03648
16 12月, 2019 3 次提交
- C
  
  mobilenet works · e9833139
  由 chonwhite 提交于 12月 16, 2019
  
  e9833139
- 石
  update profiler, test=develop (#2607) · 41826a31
  由石晓伟提交于 12月 16, 2019
```
* update profiler, test=develop

* warm up times of profiler, test=develop
```
  41826a31
- 石
  update profiler, test=develop (#2607) · af37a14f
  由石晓伟提交于 12月 16, 2019
```
* update profiler, test=develop

* warm up times of profiler, test=develop
```
  af37a14f