提交 · 0db9f7281577080821f727b0b5b13c523d2a349c · PaddlePaddle / Paddle-Lite

16 7月, 2020 4 次提交
- W
  
  [CUDA] [Kernels] Add gru fp16 cuda kernel. (#3956) · 796e2a57
  由 Wilber 提交于 7月 16, 2020
  
  796e2a57
- W
  
  [CUDA] [Kernel] Add sigmoid cuda kernel. (#3955) · 14397ca0
  由 Wilber 提交于 7月 16, 2020
  
  14397ca0
- H
  [arm]update sequence_pool and sequence_conv profiler (#3934) · bebe26e5
  由 HappyAngel 提交于 7月 16, 2020
```
* update sequence_pool and sequence_conv profiler, test=develop

* fix compute error, test=develop

* delete warning and extra info, test=develop
```
  bebe26e5
- W
  
  [CUDA] [Kernel] Add gru cuda fp32 kernel. (#3953) · 611c9b37
  由 Wilber 提交于 7月 16, 2020
  
  611c9b37
14 7月, 2020 2 次提交
- M
  add Act grad (#3923) · 845e80d9
  由 mapingshuo 提交于 7月 14, 2020
```
add act grad ops
```
  845e80d9
- Y
  [OPENCL] remove conv redundant's for opencl kernel. test=develop (#3924) · d341fccb
  由 ysh329 提交于 7月 14, 2020
```
remove conv redundant's for opencl kernel.
```
  d341fccb
13 7月, 2020 1 次提交

[LITE][XPU] Support ResnetCbam and MMDNN (#3844) · 4780849f

由 Cwndmiao 提交于 7月 13, 2020

* [LITE][XPU] accomodate resnet_cbam

* [LITE][XPU] accomodate content-dnn

* fix pr comments test=develop

* fix pr comments test=develop

* fix pr comments test=develop test=xpu

* fix compilation error, test=develop test=xpu

* [X86] Fix the unit test of slice op
test=develop test=xpu
Co-authored-by: Nhong19860320 <9973393+hong19860320@users.noreply.github.com>

4780849f

10 7月, 2020 1 次提交
- Q
  [NPU] enhance cache offline model, test=develop (#3805) · 97b54fbe
  由 Qi Li 提交于 7月 10, 2020
```
* [NPU] enhance cache offline model, test=develop
```
  97b54fbe
08 7月, 2020 3 次提交
- W
  
  [CUDA] [Kernel] Add cuda fp16 kernel (#3903) · af514209
  由 Wilber 提交于 7月 08, 2020
  
  af514209
- Y
  [OPENCL] Fix opencl fc int16 model bug caused by fc kernel (#3900) · 62c6d5d5
  由 ysh329 提交于 7月 08, 2020
```
* fix opencl fc kernel caused int16 model weight abnormal. test=develop
```
  62c6d5d5
- W
  
  [CUDA] [Kernel] Add matmul cuda kernel. (#3897) · 498f147d
  由 Wilber 提交于 7月 08, 2020
  
  498f147d
07 7月, 2020 1 次提交
- Y
  [ARM] add support ttfnet. test=develop (#3876) · 79c54557
  由 ysh329 提交于 7月 07, 2020
```
* fix ttfnet miscs. test=develop
```
  79c54557
06 7月, 2020 1 次提交
- M
  
  [MLU] add cast on MLU as default, test=develop (#3776) · cc927184
  由 MaxwellDing 提交于 7月 06, 2020
  
  cc927184
03 7月, 2020 1 次提交
- W
  
  update transpose cuda kernel. test=develop (#3879) · 11cbd50e
  由 Wilber 提交于 7月 03, 2020
  
  11cbd50e
30 6月, 2020 3 次提交
- W
  
  [Kernels] [CUDA] Add sequence_pad & unpad cuda kernel. (#3858) · 126691f0
  由 Wilber 提交于 6月 30, 2020
  
  126691f0
- W
  
  [Fluid-Lite] Remove all PADDLE_ENFORCE and PADDLE_THROW. (#3785) · ab8af5c4
  由 Wilber 提交于 6月 30, 2020
  
  ab8af5c4
- H
  [arm] update con int8 kernel choose (#3834) · c67b92b1
  由 HappyAngel 提交于 6月 29, 2020
```
* fix conv int8 kernel choose and sooftmax compute bug

* change axis_size = 4 kernel choose, test=develop

* fix format. test=develop
```
  c67b92b1
26 6月, 2020 1 次提交
- C
  
  Add clip op, test=develop (#3840) · d9c3ae01
  由 cc 提交于 6月 26, 2020
  
  d9c3ae01
23 6月, 2020 1 次提交
- W
  
  add topk_pooling kernel. test=develop (#3813) · 2402c742
  由 Wilber 提交于 6月 23, 2020
  
  2402c742
22 6月, 2020 1 次提交
- D
  
  [OPENCL]add transpose、transpose2 kernel, test=develop · 732bb91b
  由 dustybluebird 提交于 6月 22, 2020
  
  732bb91b
15 6月, 2020 6 次提交
- W
  
  support compile with cuda11 + cudnn8. test=develop (#3788) · 79a32b9d
  由 Wilber 提交于 6月 15, 2020
  
  79a32b9d
- Y
  
  [ARM] add elementwise mod operator, test=develop (#3726) · 4e9852e7
  由 yiicy 提交于 6月 15, 2020
  
  4e9852e7
- H
  [arm] add 3x3s1 Winograd int8 (#3767) · 3b4eee0f
  由 HappyAngel 提交于 6月 15, 2020
```
* fix: winograd support unsame pad
test=develop

* feat: add winograd int8 kernel
test=develop

* fix: style fix
test=develo

* fix winograd_int8 ut sgement default. test=develop

* close basic_test, test=develop
Co-authored-by: NMyPandaShaoxiang <txg4794@163.com>
```
  3b4eee0f
- Q
  
  [NPU] add thresholded_relu for arm and npu, test=develop (#3786) · 8e400754
  由 Qi Li 提交于 6月 15, 2020
  
  8e400754
- Y
  
  [LITE][OPENCL] Fix create command queue with properties. test=develop (#3772) · b081fcb0
  由 Yuan Shuai 提交于 6月 15, 2020
  
  b081fcb0
- H
  [Arm]Add Group_norm OP (#3781) · c0a3b120
  由 HappyAngel 提交于 6月 14, 2020
```
* add grouup_norm

* fix format. test=develop

* fix xiaodu crash. test=develop
```
  c0a3b120
11 6月, 2020 1 次提交
- W
  
  [CUDA] [NVTX] Lite add nvtx to support performance debug. (#3764) · 6299a90a
  由 Wilber 提交于 6月 11, 2020
  
  6299a90a
08 6月, 2020 2 次提交
- H
  [arm] fix 3x3s2p0 avg pooling crash (#3754) · 3c1d9817
  由 HappyAngel 提交于 6月 07, 2020
```
* fix pooling overflow, test=develop

* fix conflict test=develop
```
  3c1d9817
- Y
  [LITE][OPENCL] Fix grid sampler for mali gpu (#3753) · 5b091f91
  由 Yuan Shuai 提交于 6月 08, 2020
```
* [LITE][OPENCL] Fix grid sampler for mali gpu. test=develop
```
  5b091f91
05 6月, 2020 2 次提交

Y
[LITE][OEPNCL] Fix opencl compatibility error for opencl driver 1.10 (#3749) · 1c415218
由 Yuan Shuai 提交于 6月 05, 2020
```
* [LITE][OPENCL] Fix OpenCL compatibility for opencl driver 1.10. test=develop
```
1c415218

[opencl] expand opencl kernel & unit test (#3742) · fbffa674

由 xiebaiyuan 提交于 6月 05, 2020

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

* [OPENCL] develop expend opencl kernel & unit test ,test=develop

fbffa674

04 6月, 2020 1 次提交

[OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop (#3733) · e5e01469

由 xiebaiyuan 提交于 6月 04, 2020

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

* [OPENCL] develop pixel_shuffle opencl kernel & unit test ,test=develop

e5e01469

03 6月, 2020 1 次提交
- Y
  [ARM] int8 direct_conv, dw_conv add relu6 and leaky relu fusion, test=develop (#3737) · 9e361a4d
  由 yiicy 提交于 6月 03, 2020
```
int8 direct_conv, dw_conv add relu6 and leaky relu fusion
```
  9e361a4d
01 6月, 2020 2 次提交
- W
  
  [CUDA] [Framework] [FP16] Lite framework support fp16. (#3673) · c0a8e2dd
  由 Wilber 提交于 6月 01, 2020
  
  c0a8e2dd
- H
  [arm] add int8 gemm/gemv + relu6/leakyRelu fusion (#3572) · c7dd1458
  由 HappyAngel 提交于 5月 31, 2020
```
* improve 3x3s1 direct profile

* add gemv+relu6/lleakyRelu

* fix relu6 problem, test=develop

* fix format, test=develop

* add six / scale , test=develop
```
  c7dd1458
28 5月, 2020 1 次提交

[Libsize] Reduce size of dynamic library ".so" (#3717) · ec8ef528

由 T8T9 提交于 5月 28, 2020

* reduce .so size. test=develop

* compile all targets when LITE_ON_TINY_PUBLISH=OFF

* unordered_map is more convenient when key is customized class

* test=develop

ec8ef528

26 5月, 2020 1 次提交

[arm]add 2x2s2p1 pooling (#3705) · 31150e99

由 HappyAngel 提交于 5月 26, 2020

* fix pooling bug and speed

* add 2x2s2p1 pooling. test=develop

* fix conflict, test=develop

31150e99

18 5月, 2020 1 次提交

[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

17 5月, 2020 1 次提交
- H
  
  [x86][CodeStyle] Remove glog and gflag from x86 base kernel implementaion code #3635 · 2db2e2d6
  由 huzhiqiang 提交于 5月 17, 2020
  
  2db2e2d6
13 5月, 2020 1 次提交
- Z
  
  [NPU] save subgraph model cache (#3589) · 56ea5fff
  由 zhupengyang 提交于 5月 13, 2020
  
  56ea5fff