提交 · 221ae84cb07828d0bceed860ba218f63b99cf6a8 · PaddlePaddle / Paddle-Lite

16 9月, 2019 5 次提交
- M
  Merge branch 'fpga_patch' of github.com:MyPandaShaoxiang/Paddle-Lite into fpga_patch · 221ae84c
  由 MyPandaShaoxiang 提交于 9月 16, 2019
```
test=develop
```
  221ae84c
- M
  fix: fix build_fpga.sh WITH_TEST=OFF · df8c332e
  由 MyPandaShaoxiang 提交于 9月 16, 2019
```
     add build_fpga publish_inference fpga dir
     test=develop
```
  df8c332e
- M
  fix: fix build_fpga.sh WITH_TEST=OFF · 44cab39f
  由 MyPandaShaoxiang 提交于 9月 16, 2019
```
     add build_fpga publish_inference fpga dir
```
  44cab39f
- H
  
  add dynamic library for tiny and full publish (#2036) · 7e6a9030
  由 huzhiqiang 提交于 9月 16, 2019
  
  7e6a9030
- X
  
  fix math dependencies error (#2023) · 5900c784
  由 Xiaoyang LI 提交于 9月 16, 2019
  
  5900c784
13 9月, 2019 2 次提交

checkout if passes match targets and kernels, test=develop (#2035) · e27e1b08

由石晓伟提交于 9月 13, 2019

* checkout if passes match targets and kernels, test=develop

* add pass_utils, test=develop

* fix lite/core/mir/pass_registry.h, test=develop

* improve code styles, test=develop

* fix spell error, test=develop

e27e1b08

Z

Add the memory optim pass (#2018) · 89b3026c
由 Zhaolong Xing 提交于 9月 13, 2019

89b3026c

12 9月, 2019 12 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
- Y
  
  integrate model_optimize_tool compilation to build.sh (#2033) · 3d9028de
  由 Yan Chunwei 提交于 9月 12, 2019
  
  3d9028de
- Z
  fix bilinear-interp arm compute (#2029) · ca424e73
  由 zhupengyang 提交于 9月 12, 2019
```
fix bilinear-interp unit test for more cases

test=develop
```
  ca424e73
- L
  
  add matmul op kernels for asr test=develop (#2032) · b6379fed
  由 lijianshe02 提交于 9月 12, 2019
  
  b6379fed
- H
  add x86 math lstm and selected_rows test=develop (#1991) · dc27e9ff
  由 huzhiqiang 提交于 9月 12, 2019
```
add math function: lstm  and selected_rows into lite/x86/math
add selected_rows and rw_lock into lite/fluid
add lstm_cpu_kernel and  lstm_kernel into lite/x86/detail
```
  dc27e9ff
- 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
- G
  
  enable native compiling on raspberry pi and rk3399 (#2021) · e0b4b5c9
  由 guofei 提交于 9月 12, 2019
  
  e0b4b5c9
- W
  add unsqueeze and range op (x2paddle) (#1988) · 3c08f676
  由 Wilber 提交于 9月 12, 2019
```
* add unsqueeze and range op. modify concat op test=develop

* modify exception in range_test_x86
```
  3c08f676
- W
  
  add min_max_aspect_ratios_order attr in prior box op test=develop (#2016) · cf84d42b
  由 Wilber 提交于 9月 12, 2019
  
  cf84d42b
- L
  
  add elementwise op function and add elementwise add/sub kernels test=develop (#2020) · 806ba6e7
  由 lijianshe02 提交于 9月 12, 2019
  
  806ba6e7
- W
  add transpose kernel for cuda test=develop (#1997) · cba5736f
  由 Wilber 提交于 9月 12, 2019
```
add transpose kernel for cuda
```
  cba5736f
11 9月, 2019 5 次提交
- Y
  
  make model_optimize_tool run on host (#1990) · 83d4b0e8
  由 Yan Chunwei 提交于 9月 11, 2019
  
  83d4b0e8
- S
  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
- 石
  make passes related to the device type, test=develop (#2012) · 8ca10db8
  由石晓伟提交于 9月 11, 2019
```
* make passes related to the device type, test=develop

* improve tips, test=develop
```
  8ca10db8
- L
  add slice op, reshape op, reshape2 op, squeeze op, squeeze2 op for x86 (#2005) · 13bbd2b8
  由 liu zhengxi 提交于 9月 11, 2019
```
add slice op, reshape op,  reshape2 op, squeeze op and squeeze2 op and their unittests for x86
```
  13bbd2b8
- Z
  fix conv-act-fuse-pass when there is no "bias" (#2003) · 2dcff5ca
  由 zhupengyang 提交于 9月 11, 2019
```
test=develop
```
  2dcff5ca
10 9月, 2019 7 次提交
- L
  
  add x86 softmax kernel and fix jit compute bugs test=develop (#2007) · 81132a32
  由 lijianshe02 提交于 9月 10, 2019
  
  81132a32
- W
  
  add elementwise_sub and modify argmax (#1964) · 62ea82d0
  由 Wilber 提交于 9月 10, 2019
  
  62ea82d0
- J
  Modify detection test (#2000) · 111db475
  由 juncaipeng 提交于 9月 10, 2019
```
* add assign_value op, arm kernel and test, add fluid_type, test=develop

* add hard_sigmoid, test=develop

* use image and new imple to test detection model, delete faster_rcnn_test, test=develop
```
  111db475
- X
  fix model_optimize_tool error when using host kernel, fix reshape op build... · f25a4571
  由 Xiaoyang LI 提交于 9月 10, 2019
```
fix model_optimize_tool error when using host kernel, fix reshape op build error on ios, test=develop (#1984)
```
  f25a4571
- T
  
  fix fpga compile problem and kernels (#1989) · 0720653b
  由 TianXiaogang 提交于 9月 10, 2019
  
  0720653b
- 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 5 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
- N
  
  fusion instancenorm and relu. test=develop (#1986) · eabee4dc
  由 NazgulLee 提交于 9月 09, 2019
  
  eabee4dc
- J
  add assign_value and hard_sigmoid, add fluid_type (#1983) · 92eeabeb
  由 juncaipeng 提交于 9月 09, 2019
```
* add assign_value op, arm kernel and test, add fluid_type, test=develop

* add hard_sigmoid, test=develop
```
  92eeabeb
- P
  Add concat and elementwise_add cuda kernel (#1979) · 6d1da405
  由 Pei Yang 提交于 9月 09, 2019
```
* add nearest_interp_cuda kernel, test=develop

* add concat op and elementwise_add op

* remove eigen dependency from nearest_interp cuda kernel, test=develop

* free cuda pointers, test=develop
```
  6d1da405
- Z
  add calib cuda kernel. (#1977) · da328594
  由 Zhen Wang 提交于 9月 09, 2019
```
* add calib cuda kernel.

* add unit test for calib cuda kernel. test=develop
```
  da328594
07 9月, 2019 1 次提交

add lite x86 ops for ASR test=develop (#1981) · 7014a76b

由 lijianshe02 提交于 9月 07, 2019

* add lite x86 ops for ASR test=develop

* add lite x86 ops for ASR test=develop

* fix x86 ci run test problems test=develop

* fix mkl path for CI test=develop

7014a76b

06 9月, 2019 3 次提交
- Y
  fix ios CPU preprocess test=develop (#1972) · 04f4775b
  由 Yanzhan Yang 提交于 9月 06, 2019
```
* fix ios CPU preprocess test=develop

* fix style test=develop
```
  04f4775b
- Z
  add interpolate fuse pass (#1980) · c49958a2
  由 zhupengyang 提交于 9月 06, 2019
```
test=develop
```
  c49958a2
- H
  modify reshape2 OP test=dvelop (#1963) · febfd7d6
  由 huzhiqiang 提交于 9月 06, 2019
```
modify reshape2 OP to add shape_tensor input
```
  febfd7d6