提交 · e3caf9dddceea92358fcf3be6d32a0375e280d7f · PaddlePaddle / Paddle-Lite

04 11月, 2019 1 次提交
- X
  
  use shared_ptr to wrap shared cl mem ,test=develop (#2364) · e3caf9dd
  由 xiebaiyuan 提交于 11月 04, 2019
  
  e3caf9dd
02 11月, 2019 1 次提交

fix useless init empty image ,test=develop (#2354) · f8fb5052

由 xiebaiyuan 提交于 11月 02, 2019

* add mem opt option for high api ,test = develop

* fix is_lod params && test-performance ,test=develop

* fix useless init emptyimage , test = develop

f8fb5052

01 11月, 2019 3 次提交
- Y
  checkout specific commit for opencl headers (#2341) · 2aa7b06b
  由 Yanzhan Yang 提交于 11月 01, 2019
```
* delete remaining opencl headers test=develop

* checkout specific commit for opencl headers test=develop
```
  2aa7b06b
- Y
  add opencl headers by git clone test=develop (#2340) · cec4efd8
  由 Yanzhan Yang 提交于 11月 01, 2019
```
* add opencl headers by git clone test=develop

* add opencl headers by git clone test=develop
```
  cec4efd8
- X
  add mem opt option for high api ,test = develop (#2319) · 624fa8d4
  由 xiebaiyuan 提交于 11月 01, 2019
```
* add mem opt option for high api ,test = develop

* fix is_lod params && test-performance ,test=develop
```
  624fa8d4
31 10月, 2019 1 次提交
- J
  add GPU_CL for test_net_benchmark. (#2307) · 2c380625
  由 Jiaying Zhao 提交于 10月 31, 2019
```
* add GPU_CL for test_net_benchmark.

* test compile

* fix mobile build error
```
  2c380625
24 10月, 2019 2 次提交
- X
  
  fix is_lod params ,test=develop (#2258) · 71684f98
  由 xiebaiyuan 提交于 10月 24, 2019
  
  71684f98
- X
  Develop performance (#2257) · f0f3e90b
  由 xiebaiyuan 提交于 10月 24, 2019
```
* optimise conv 1x1 ,test=develop

* add test_performance test
```
  f0f3e90b
23 10月, 2019 2 次提交
- X
  
  optimise conv 1x1 ,test=develop (#2248) · b382a0dd
  由 xiebaiyuan 提交于 10月 23, 2019
  
  b382a0dd
- X
  
  fix out range in conv 1x1 for nano yolo ,test=develop (#2246) · 61a89ef2
  由 xiebaiyuan 提交于 10月 23, 2019
  
  61a89ef2
22 10月, 2019 1 次提交
- Y
  
  handle nan in run.py test=develop (#2240) · cbe7098e
  由 Yanzhan Yang 提交于 10月 22, 2019
  
  cbe7098e
21 10月, 2019 1 次提交
- J
  
  add GetExceptionMsg for paddle_inference_api. test=develop (#2231) · fda4d42c
  由 Jiaying Zhao 提交于 10月 21, 2019
  
  fda4d42c
17 10月, 2019 1 次提交
- S
  
  fix “CL_INVALID_KERNEL_ARGS ” error， test=develop (#2213) · 2667a153
  由 StarryRain 提交于 10月 17, 2019
  
  2667a153
16 10月, 2019 2 次提交
- J
  
  Open merge_cl_to_so switch and delete -I(cl_path) build option. test=develop (#2206) · 781d8191
  由 Jiaying Zhao 提交于 10月 16, 2019
  
  781d8191
- X
  
  support global pooling ... test=develop (#2204) · b963383a
  由 xiebaiyuan 提交于 10月 16, 2019
  
  b963383a
15 10月, 2019 1 次提交
- Y
  
  fix persistable test=develop (#2191) · 435f942b
  由 Yanzhan Yang 提交于 10月 15, 2019
  
  435f942b
12 10月, 2019 1 次提交
- J
  
  fix clang compile error. test=develop (#2180) · 508ca98b
  由 Jiaying Zhao 提交于 10月 12, 2019
  
  508ca98b
11 10月, 2019 1 次提交
- Y
  
  1. fix group logic for convolution op. 2. add pixel shuffle op for OpenCL. (#2178) · 77811367
  由 Yanzhan Yang 提交于 10月 11, 2019
  
  77811367
10 10月, 2019 2 次提交
- Y
  1. improve n-fold quantification algorithm by introducing a minimal size for... · 4bad9853
  由 Yanzhan Yang 提交于 10月 10, 2019
```
1. improve n-fold quantification algorithm by introducing a minimal size for each fold. 2. automatically search for best n for n-fold algorithm. (#2167)
```
  4bad9853
- X
  
  fix an calc bug in test-mobilenetgpu (#2162) · f4a42296
  由 xiebaiyuan 提交于 10月 10, 2019
  
  f4a42296
09 10月, 2019 4 次提交
- Y
  add n-fold quantification algorithm (#2164) · 7e9bb98a
  由 Yanzhan Yang 提交于 10月 09, 2019
```
* 1. add quantification_fold parameter. 2. support quantification test in run.py.

* implement n-fold quantification
```
  7e9bb98a
- X
  open cl mem optimise, split with cpu codes. fix a bug when some memory is not equal 4 . (#2161) · 33c335a5
  由 xiebaiyuan 提交于 10月 09, 2019
```
* open cl mem optimise, split with cpu codes.
fix a bug when some memory is not equal 4 .

* open cl mem optimise, split with cpu codes.   fix a bug when some memory is not equal 4  fix bad des
```
  33c335a5
- Z
  
  fix crash in some platform (#2159) · c551e1d6
  由 zp7 提交于 10月 09, 2019
  
  c551e1d6
- X
  
  paddle mobile runtime cl memory optimise. test=develop (#2160) · 528fd741
  由 xiebaiyuan 提交于 10月 09, 2019
  
  528fd741
08 10月, 2019 1 次提交
- J
  
  add pre and post process in feed and fetch kernel test=develop (#2157) · 5f227934
  由 Jiaying Zhao 提交于 10月 08, 2019
  
  5f227934
27 9月, 2019 2 次提交
- Z
  
  set merge_cl_to_so default value 0 (#2145) · ff00083d
  由 zp7 提交于 9月 27, 2019
  
  ff00083d
- N
  
  optimize instance norm. test=develop (#2120) · a96d013c
  由 NazgulLee 提交于 9月 27, 2019
  
  a96d013c
26 9月, 2019 1 次提交
- Z
  
  fix crash (#2139) · 4139dc2c
  由 zp7 提交于 9月 26, 2019
  
  4139dc2c
25 9月, 2019 1 次提交

get opencl kernel_work_group_size (#2123) · 0c51266a

由 StarryRain 提交于 9月 25, 2019

* get opencl kernel_work_group_size and set local_work_group_size for conv and conv_trans kernel of GAN, test=develop

* fix fpga CI error, test=develop

* fix fpga CI error, test=develop

* auto set local work group size according to devices

* Optimal OPENCL conv_trans_7x7 Performance , test=develop

* fix CI error, test=develop

0c51266a

24 9月, 2019 1 次提交
- J
  fix fpga ci build error (#2127) · dd0c0595
  由 Jiaying Zhao 提交于 9月 24, 2019
```
* fix fpga ci build error. test=develop

* fix fpga ci build error. test=develop
```
  dd0c0595
17 9月, 2019 2 次提交
- Y
  fix new_scale & new_bias crash test=develop (#2055) · 90ddbd63
  由 Yanzhan Yang 提交于 9月 17, 2019
```
* fix new_scale & new_bias crash test=develop

* fix new_bias & new_scale in GPU test=develop
```
  90ddbd63
- Y
  
  add assign_value op for gpu test=develop (#2053) · 4aba7e55
  由 Yanzhan Yang 提交于 9月 17, 2019
  
  4aba7e55
16 9月, 2019 1 次提交
- Y
  
  refine paddle_inference_api.h test=develop (#2048) · a2a5c8b0
  由 Yanzhan Yang 提交于 9月 16, 2019
  
  a2a5c8b0
12 9月, 2019 3 次提交
- S
  add opencl conv_trans_op (#2031) · 3d59e558
  由 StarryRain 提交于 9月 12, 2019
```
* add opencl  depthwise_conv_trans_op

* test=develop

* Optimal OPENCL conv_3x3S2 Performance

* test=develop

* Optimal OPENCL conv_trans_3x3s2 Performance , test=develop

* add opencl conv_trans_op, test=develop
```
  3d59e558
- N
  
  optimize conv7*7, compute 2 point in 1 thread. test=develop (#2028) · 9851c008
  由 NazgulLee 提交于 9月 12, 2019
  
  9851c008
- Z
  
  fix gpu load memory leak,test=develop (#2027) · 8ab543ff
  由 zp7 提交于 9月 12, 2019
  
  8ab543ff
11 9月, 2019 1 次提交

Optimal OPENCL conv_trans_3x3s2 Performance (#2015) · 487d6089

由 StarryRain 提交于 9月 11, 2019

* add opencl  depthwise_conv_trans_op

* test=develop

* Optimal OPENCL conv_3x3S2 Performance

* test=develop

* Optimal OPENCL conv_trans_3x3s2 Performance , test=develop

487d6089

10 9月, 2019 2 次提交
- N
  
  optimize instancenorm local workgroup computation. test=develop (#2001) · 019f5b8e
  由 NazgulLee 提交于 9月 10, 2019
  
  019f5b8e
- S
  Optimal OPENCL conv_3x3 Performance (#1994) · b50f52cd
  由 StarryRain 提交于 9月 10, 2019
```
* add opencl  depthwise_conv_trans_op

* test=develop

* Optimal OPENCL conv_3x3S2 Performance

* test=develop
```
  b50f52cd
09 9月, 2019 1 次提交
- Y
  fix vector out of range crash test=develop (#1999) · 58ab6c29
  由 Yanzhan Yang 提交于 9月 09, 2019
```
* fix vector out of range crash test=develop

* fix style test=develop
```
  58ab6c29