提交 · 425edbfb5d546c403189216d6b049ae84529aeb4 · PaddlePaddle / Paddle-Lite

09 1月, 2020 1 次提交
- X
  
  fix: fix 5x5 choose kernel bug (#2741) (#2744) · 425edbfb
  由 xiaogang 提交于 1月 09, 2020
  
  425edbfb
31 12月, 2019 2 次提交
- J
  fix mask rcnn error when run twice, test=develop (#2675) (#2709) · cb246fc8
  由 juncaipeng 提交于 12月 31, 2019
```
add clear for tensor
```
  cb246fc8
- J
  Fix yolo_box bug (#2704) (#2707) · d56f25df
  由 juncaipeng 提交于 12月 31, 2019
```
* fix yolov3 bug when run several times, test=develop
```
  d56f25df
30 12月, 2019 1 次提交

fix op inputs and outputs type (#2647) (#2702) · f1bc89c7

由 huzhiqiang 提交于 12月 30, 2019

* fix op inputs and outputs type, test=develop
Co-authored-by: Njuncaipeng <52520497+juncaipeng@users.noreply.github.com>

f1bc89c7

27 12月, 2019 1 次提交
- H
  
  add arm kernels into model_opt_tool (#2692) · c01151d3
  由 huzhiqiang 提交于 12月 27, 2019
  
  c01151d3
25 12月, 2019 1 次提交
- Y
  
  [ARM] multiclass_nms op add index output, test=develop (#2654) (#2667) · 58a18824
  由 yiicy 提交于 12月 25, 2019
  
  58a18824
23 12月, 2019 2 次提交
- W
  add sequence_pool_concat fuse and kernel test=develop (#2645) · 1b74fded
  由 Wilber 提交于 12月 23, 2019
```
add sequence_pool_concat fuse pass

add fuse kernel
```
  1b74fded
- Y
  
  [ARM] add grid_sampler op and ut, test=develop (#2598) · 3723451b
  由 yiicy 提交于 12月 23, 2019
  
  3723451b
21 12月, 2019 1 次提交
- Z
  [XPU] add layer_norm bridge and unit test (#2640) · 4dd6a4b8
  由 zhupengyang 提交于 12月 21, 2019
```
test=develop
```
  4dd6a4b8
20 12月, 2019 3 次提交
- Z
  [XPU] add reshape bridge and unit test (#2621) · a13c592d
  由 zhupengyang 提交于 12月 20, 2019
```
test=develop
```
  a13c592d
- W
  add var_conv_2d_relu pass test=develop (#2631) · 8304bc84
  由 Wilber 提交于 12月 20, 2019
```
add var_conv_2d + relu fuse pass
```
  8304bc84
- Z
  [XPU] add transpose bridge and unit test (#2630) · b53ece7a
  由 zhupengyang 提交于 12月 20, 2019
```
* [XPU] add transpose bridge and unit test

test=develop
```
  b53ece7a
19 12月, 2019 3 次提交
- W
  optimize cuda kernel test=develop (#2628) · 09aa15a5
  由 Wilber 提交于 12月 19, 2019
```
* optimize content-dnn cuda kernel
```
  09aa15a5
- X
  feature: update fpga kernel patch (#2627) · 3f35879b
  由 xiaogang 提交于 12月 19, 2019
```
* feature: update fpga kernel patch
```
  3f35879b
- 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
18 12月, 2019 1 次提交
- J
  Support Mask RCNN2 (#2588) · d1b7aec5
  由 juncaipeng 提交于 12月 18, 2019
```
* Support Mask RCNN2 (#2588)
```
  d1b7aec5
17 12月, 2019 3 次提交
- H
  
  [lite][arm] add conv+relu6/leakyRelu fusion (#2599) · 3455ab0a
  由 HappyAngel 提交于 12月 17, 2019
  
  3455ab0a
- X
  Develop lite reshape (#2613) · e3808d79
  由 xiebaiyuan 提交于 12月 17, 2019
```
* add reshape opencl kernel && optimise conv 1x1 ,test=develop

* add reshape opencl kernel && optimise conv 1x1 &&code style ,test=develop

* add reshape opencl kernel && optimise conv 1x1 &&code style ,test=develop
```
  e3808d79
- Y
  
  add fp16 opencl kernel. fix nhwc as imagedefault for opencl layout. test=develop (#2616) · 24c98e60
  由 Yuan Shuai 提交于 12月 17, 2019
  
  24c98e60
16 12月, 2019 2 次提交

[LITE][OPENCL] Add relu image2d kernel unit test, Fix conv2d_1x1, relu, layout... · ee177c6b

由 Yuan Shuai 提交于 12月 16, 2019

[LITE][OPENCL] Add relu image2d kernel unit test, Fix conv2d_1x1, relu, layout using new Image2D Layout (#2564)

* add 3 layout for opencl image. test=develop

* add relu image2d test. test=develop

ee177c6b

[LITE][OPENCL] Add depthwise_conv_3x3 opencl kernel (#2601) · 600c8c20

由 Jiaying Zhao 提交于 12月 16, 2019

* [LITE][OPENCL] Add depthwise_conv_3x3 opencl kernel

* [LITE][OPENCL] Add depthwise_conv_3x3 opencl kernel. test=develop

* [LITE][OPENCL] Add Pool opencl kernel. test=develop

600c8c20

15 12月, 2019 1 次提交
- W
  optimize search_grnn test=develop (#2608) · dad43f81
  由 Wilber 提交于 12月 15, 2019
```
optimize search_grnn
```
  dad43f81
13 12月, 2019 1 次提交
- H
  [LITE][NPU][XPU] Refine subgraph pass, and support NPU/XPU model generation at... · d5434aa2
  由 hong19860320 提交于 12月 13, 2019
```
[LITE][NPU][XPU] Refine subgraph pass, and support NPU/XPU model generation at execution time (#2576)
```
  d5434aa2
12 12月, 2019 1 次提交

[LITE][OPENCL] Add conv2d_1x1 opencl kernel (#2591) · e583a55d

由 xiebaiyuan 提交于 12月 12, 2019

* add opencl conv1x1 image impl and unit test pass with relu & bias,
add layout_compute --> buffer2image float32 --> with unit test pass
suite checked test for more situation , test=develop

* add opencl conv1x1 image impl and unit test pass with relu & bias,
add layout_compute --> buffer2image float32 --> with unit test pass
suite checked test for more situation , test=develop

* fix white space cpp lint , test=develop

e583a55d

11 12月, 2019 1 次提交
- T
  
  add winograd f23 implement (#2584) · f99c34c8
  由 TianXiaogang 提交于 12月 11, 2019
  
  f99c34c8
10 12月, 2019 3 次提交
- H
  
  modify code_style of CMakeList.txt to make ci_build_test_server able to work on gcc_4.8.2 · 4ae43560
  由 huzhiqiang 提交于 12月 10, 2019
  
  4ae43560
- Y
  
  [ARM] add instance norm op and ut, test=develop (#2578) · 9a3552db
  由 yiicy 提交于 12月 10, 2019
  
  9a3552db
- Z
  [NPU] add argamx op bridge and unit test (#2580) · c64036ca
  由 zhupengyang 提交于 12月 10, 2019
```
test=develop
```
  c64036ca
09 12月, 2019 1 次提交
- Z
  [NPU] support relu6 (#2582) · a2f981a4
  由 zhupengyang 提交于 12月 09, 2019
```
test=develop
```
  a2f981a4
08 12月, 2019 1 次提交
- L
  
  Add fc op on lite x86 platform (#2568) · d76c529a
  由 liu zhengxi 提交于 12月 08, 2019
  
  d76c529a
07 12月, 2019 2 次提交

Z
[NPU] add unsqueeze op bridge and unit test (#2570) · 3f28e00e
由 zhupengyang 提交于 12月 07, 2019
```
test=develop
```
3f28e00e

Support mask_rcnn (#2484) · c2f72cb3

由 juncaipeng 提交于 12月 07, 2019

* 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

06 12月, 2019 1 次提交
- W
  fix fill_constant bug and add int64->int32 cast test=develop (#2566) · e17295cc
  由 Wilber 提交于 12月 06, 2019
```
- fix fill_constant bug.
- cast op support int64_t->int32_t
```
  e17295cc
04 12月, 2019 2 次提交
- W
  update cuda kernels to run content-dnn models test=develop (#2554) · aa67c28e
  由 Wilber 提交于 12月 04, 2019
```
update cuda kernels to run content-dnn model
```
  aa67c28e
- Z
  [cuda] [int8] resnet50 cuda int8 support (#2417) · f7574646
  由 Zhaolong Xing 提交于 12月 04, 2019
```
* init resnet cuda int8 support
test=develop

* refine cuda unit test
test=develop

* add the forgeted file.
test=develop
```
  f7574646
03 12月, 2019 2 次提交
- J
  set the bias of int8 conv as float, test=develop (#2553) · 04d2b4eb
  由 juncaipeng 提交于 12月 03, 2019
```
* set the bias of int8 conv as float, test=develop
```
  04d2b4eb
- Z
  [NPU] support 1-dimension input y in elemetwise ops (#2546) · 919282cf
  由 zhupengyang 提交于 12月 03, 2019
```
test=develop
```
  919282cf
02 12月, 2019 2 次提交
- L
  
  delete useless code for x86 platform (#2535) · 1b875ae8
  由 liu zhengxi 提交于 12月 02, 2019
  
  1b875ae8
- L
  
  Fix the conv compute shape (#2534) · 0a100120
  由 liu zhengxi 提交于 12月 02, 2019
  
  0a100120
30 11月, 2019 1 次提交
- Z
  [NPU] fix act; refine act unit tests; fix batch_norm (#2533) · 0349bfd0
  由 zhupengyang 提交于 11月 30, 2019
```
test=develop
```
  0349bfd0