提交 · fea601e92c5709d1ca18313a35d2eaf8a38c8a77 · PaddlePaddle / Paddle-Lite

22 5月, 2019 1 次提交
- J
  add opencl Scale op (#1645) · fea601e9
  由 Jiaying Zhao 提交于 5月 22, 2019
```
* fix memcpy size in opencl fetch kernel

* add opencl Scale op
```
  fea601e9
21 5月, 2019 6 次提交
- Y
  1.add groups to conv3x3. 2.add original size to tensor and tenture and adjust... · 02ebdf59
  由 Yanzhan Yang 提交于 5月 21, 2019
```
1.add groups to conv3x3. 2.add original size to tensor and tenture and adjust concat layer to handle tensor truncation logic. (#1643)
```
  02ebdf59
- Add leakyrelu arm cpu op close #1636 (#1638) · e8200397
  由 xiebaiyuan 提交于 5月 21, 2019
  
  e8200397
- N
  
  method to adjust texture transpose (#1642) · bd961137
  由 NazgulLee 提交于 5月 21, 2019
  
  bd961137
- N
  
  add name protection for protobuf-c (#1641) · 03860d43
  由 NazgulLee 提交于 5月 21, 2019
  
  03860d43
- Y
  
  1.check slice and nearest op result. 2.reshape output should have same layout as its input. (#1640) · 5a351346
  由 Yanzhan Yang 提交于 5月 21, 2019
  
  5a351346
- J
  opendevice bug fixed for FPGA V1 track (#1635) · f7ebe337
  由 jameswu2014 提交于 5月 21, 2019
```
* V2-conv-hellocase pass & V1 verify-pass

* V1 paralvl equal to 2

* softmax channel =2 bug

* softmaxbug update2

* io update2

* V1-dwconv relu bug

* opendevice bug fixed
```
  f7ebe337
20 5月, 2019 2 次提交
- Y
  1.add flatten2, slice and nearest_interp op. 2.truncate tensor with more than... · 5fddcef8
  由 Yanzhan Yang 提交于 5月 20, 2019
```
1.add flatten2, slice and nearest_interp op. 2.truncate tensor with more than 4 dimensions to 4. (#1634)
```
  5fddcef8
- Y
  
  add exp, sigmoid and leaky_relu op (#1633) · 96e616c5
  由 Yanzhan Yang 提交于 5月 20, 2019
  
  96e616c5
18 5月, 2019 1 次提交
- Y
  1.parse tensor desc info from params file. 2.ensure tensor desc is consistent... · bc9875b8
  由 Yanzhan Yang 提交于 5月 18, 2019
```
1.parse tensor desc info from params file. 2.ensure tensor desc is consistent between params file and model file. (#1631)
```
  bc9875b8
17 5月, 2019 2 次提交
- J
  
  fix memcpy size in opencl fetch kernel (#1630) · 7a4e06d3
  由 Jiaying Zhao 提交于 5月 17, 2019
  
  7a4e06d3
- add notice for android gpu doc (#1627) · d7f92843
  由 xiebaiyuan 提交于 5月 17, 2019
  
  d7f92843
16 5月, 2019 3 次提交
- N
  
  add reshape2, transpose2, relu6 and scale operator (#1628) · f253fbf9
  由 NazgulLee 提交于 5月 16, 2019
  
  f253fbf9
- X
  
  fix crash when close optimise in gpu test close #1626 · d3f88584
  由 xiebaiyuan 提交于 5月 16, 2019
  
  d3f88584
- J
  V1-dwconv relu bug for FPGA track (#1624) · de71b46d
  由 jameswu2014 提交于 5月 16, 2019
```
* V2-conv-hellocase pass & V1 verify-pass

* V1 paralvl equal to 2

* softmax channel =2 bug

* softmaxbug update2

* io update2

* V1-dwconv relu bug
```
  de71b46d
15 5月, 2019 2 次提交

Y

add nslock to load and clear method of runner (#1623) · 4cf9bb15
由 Yanzhan Yang 提交于 5月 15, 2019

4cf9bb15

remove dafault opencl kernel path (#1622) · 5282c65d

由 Jiaying Zhao 提交于 5月 15, 2019

* optimize GPU conv performance and structure

* add CL macro for test_conv_gpu

* fix build failure

* change funtion name

* change funtion name

* fix load_when_predict mode in opencl

* remove dafault opencl kernel path

5282c65d

13 5月, 2019 3 次提交

fix load_when_predict mode in opencl (#1619) · c6d366c2

由 Jiaying Zhao 提交于 5月 13, 2019

* optimize GPU conv performance and structure

* add CL macro for test_conv_gpu

* fix build failure

* change funtion name

* change funtion name

* fix load_when_predict mode in opencl

c6d366c2

update feed fetch and softmax kernel in FPGA v2 track fixed#1615 (#1616) · cf542116

由 qnqinan 提交于 5月 13, 2019

* update concat and split kernel and related files in FPGA v2(v3) track

* update

* update

* update kernel and related files in FPGA v2 track

* update

* update

* update kernel and related files for static quantization in FPGA v2 track

* update

* update feed and fetch kernel in FPGA v2 track

* update io file

* update feed fetch and softmax kernel in FPGA v2 track

cf542116

S
Update ocr test case (#1605) · 3a9016d7
由 Shuai Yuan 提交于 5月 13, 2019
```
* Update ocr test case

* format code

* format code
```
3a9016d7

10 5月, 2019 4 次提交
- Y
  
  remove thread unsafe dictionary (#1617) · 9070d4cf
  由 Yanzhan Yang 提交于 5月 10, 2019
  
  9070d4cf
- Y
  
  change pixelFormat only when useAggressiveOptimization is set (#1614) · 3826f160
  由 Yanzhan Yang 提交于 5月 10, 2019
  
  3826f160
- S
  
  Add comment for ARM Linux doc (#1604) · ff21cde5
  由 Shuai Yuan 提交于 5月 10, 2019
  
  ff21cde5
- J
  io+softmax update for FPGA V1 track (#1613) · 542008c0
  由 jameswu2014 提交于 5月 10, 2019
```
* V2-conv-hellocase pass & V1 verify-pass

* V1 paralvl equal to 2

* softmax channel =2 bug

* softmaxbug update2

* io update2
```
  542008c0
09 5月, 2019 2 次提交
- Y
  
  add useAggresiveOptimization param to control newly added optimization code branch (#1612) · c014bdbf
  由 Yanzhan Yang 提交于 5月 09, 2019
  
  c014bdbf
- J
  support V1 para equal to 2 (#1609) · 04610db8
  由 jameswu2014 提交于 5月 09, 2019
```
* V2-conv-hellocase pass & V1 verify-pass

* V1 paralvl equal to 2
```
  04610db8
08 5月, 2019 3 次提交

J

V2-conv-hellocase pass & V1 verify-pass (#1608) · 49038a6a
由 jameswu2014 提交于 5月 08, 2019

49038a6a
Y
shift filter conversion process to loading phase (#1606) · e46fa929
由 Yanzhan Yang 提交于 5月 08, 2019
```
* shift filter conversion process to loading phase

* fix winograd edge case
```
e46fa929

update io file to surport int8 tensor feed and fetch fixed#1602 (#1603) · ee4b392a

由 qnqinan 提交于 5月 08, 2019

* update concat and split kernel and related files in FPGA v2(v3) track

* update

* update

* update kernel and related files in FPGA v2 track

* update

* update

* update kernel and related files for static quantization in FPGA v2 track

* update

* update feed and fetch kernel in FPGA v2 track

* update io file

ee4b392a

07 5月, 2019 3 次提交

update feed and fetch kernel in FPGA v2 track fixed#1600 (#1601) · 4c7352b1

由 qnqinan 提交于 5月 07, 2019

* update concat and split kernel and related files in FPGA v2(v3) track

* update

* update

* update kernel and related files in FPGA v2 track

* update

* update

* update kernel and related files for static quantization in FPGA v2 track

* update

* update feed and fetch kernel in FPGA v2 track

4c7352b1

S
Add readme for image preprocess and fix bug (#1597) · 77041ac4
由 Shuai Yuan 提交于 5月 07, 2019
```
* add comments for imagetools

* format code
```
77041ac4

change funtion name (#1598) · 98ab80cd

由 Jiaying Zhao 提交于 5月 07, 2019

* optimize GPU conv performance and structure

* add CL macro for test_conv_gpu

* fix build failure

* change funtion name

* change funtion name

98ab80cd

06 5月, 2019 2 次提交
- Y
  implement winograd for metal depthwise 3x3 naively (#1596) · e0ce2b3c
  由 Yanzhan Yang 提交于 5月 06, 2019
```
* implement winograd for metal depthwise 3x3 naively

* keep half4 in non-winograd depthwise 3x3 to ensure precision
```
  e0ce2b3c
- J
  optimize GPU conv performance and structure (#1595) · cf70102b
  由 Jiaying Zhao 提交于 5月 06, 2019
```
* optimize GPU conv performance and structure

* add CL macro for test_conv_gpu

* fix build failure
```
  cf70102b
30 4月, 2019 4 次提交
- Y
  
  enable all mps conv layer (#1592) · 619acd23
  由 Yanzhan Yang 提交于 4月 30, 2019
  
  619acd23
- N
  
  fix semaphore crash (#1591) · e7328579
  由 NazgulLee 提交于 4月 30, 2019
  
  e7328579
- update kernel and related files for static quantization in FPGA v2 track fixed#1589 (#1590) · ceda3f4e
  由 qnqinan 提交于 4月 30, 2019
```
* update concat and split kernel and related files in FPGA v2(v3) track

* update

* update

* update kernel and related files in FPGA v2 track

* update

* update

* update kernel and related files for static quantization in FPGA v2 track

* update
```
  ceda3f4e
- S
  Update Build docs for ARM Linux CPU and GPU(OpenCL) (#1583) · 15794c0a
  由 Shuai Yuan 提交于 4月 30, 2019
```
* format CMakeLists

* BugFix: lacking of header file of memset, strlen, stable_sort, n_copy

* Update ARM Linux CPU, GPU build docs for rk3399
```
  15794c0a
29 4月, 2019 2 次提交
- Y
  fuse Conv-Add-Relu into one op (#1588) · 5a6bfa4a
  由 Yanzhan Yang 提交于 4月 29, 2019
```
* fuse Conv-Add-Relu into one op.

* Revert "fuse Conv-Add-Relu into one op."

This reverts commit 200be579.

* fuse Conv-Add-Relu into one op
```
  5a6bfa4a
- W
  
  initTexture interface modeify to throw an exception (#1587) · 033bda03
  由 WayneLiu 提交于 4月 29, 2019
  
  033bda03