KSNN使用说明

这篇文档主要介绍什么是KSNN以及如何在VIM3上面运行提供的示例。

安装KSNN

  1. clone代码到本地
khadas@Khadas:~$ git clone --recursive https://github.com/khadas/ksnn.git
  1. 安装依赖
khadas@Khadas:~$ pip3 install matplotlib
  1. 安装KSNN库
khadas@Khadas:~$ cd ksnn/ksnn
khadas@Khadas:~/ksnn/ksnn$ pip3 install ksnn-1.3-py3-none-any.whl

使用示例

demo全部集中在examlpes目录下，

khadas@Khadas:~$ cd ksnn/examples/ && ls
caffe  darknet  keras  onnx  pytorch  tensorflow  tflite

这里以Inception V3为例，其他demo是类似的。

khadas@Khadas:~/ksnn/examples$ cd tensorflow && ls
README.md  box_priors.txt  data  inceptionv3.py  libs  mobilenet_ssd_picture.py  models

运行的命令和转换参数都在对应目录下的README文件里。

khadas@Khadas:~/ksnn/examples/tensorflow$ cat README.md
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
# run

$ python3 inceptionv3.py --model ./models/VIM3/inceptionv3.nb --library ./libs/libnn_inceptionv3.so --picture ./data/goldfish_299x299.jpg --level 0
$ python3 mobilenet_ssd_picture.py --model ./models/VIM3/mobilenet_ssd.nb --library ./libs/libnn_mobilenet_ssd.so --picture data/1080p.bmp --level 0

# Convert

$ ./convert \
--model-name inception \
--platform tensorflow \
--model inception_v3_2016_08_28_frozen.pb \
--input-size-list '299,299,3' \
--inputs input \
--outputs InceptionV3/Predictions/Reshape_1 \
--mean-values '128,128,128,128' \
--quantized-dtype asymmetric_affine \
--kboard VIM3 --print-level 1

$ ./convert \
--model-name mobilenet_ssd \
--platform tensorflow \
--model ssd_mobilenet_v1_coco_2017_11_17.pb \
--input-size-list '300,300,3' \
--inputs FeatureExtractor/MobilenetV1/MobilenetV1/Conv2d_0/BatchNorm/batchnorm/mul_1 \
--outputs "'concat concat_1'" \
--mean-values '127.5,127.5,127.5,127.5' \
--quantized-dtype asymmetric_affine \
--kboard VIM3 --print-level 1

If you use VIM3L , please use `VIM3L` to replace `VIM3`

运行Inception V3.

khadas@Khadas:~/ksnn/examples/tensorflow$ python3 inceptionv3.py --model ./models/VIM3/inceptionv3.nb --library ./libs/libnn_inceptionv3.so --picture ./data/goldfish_299x299.jpg --level 0
1
2
3
4
5
6
7
8
9
10
11
12
13
 |--- KSNN Version: v1.3 +---| 
Start init neural network ...
Done.
Get input data ...
Done.
Start inference ...
Done. inference :  0.042353153228759766
----- Show Top5 +-----
     2: 0.93457
   795: 0.00328
   408: 0.00158
   974: 0.00148
   393: 0.00093

--level参数可同于调整打印信息等级。下面的命令将打印等级设置为最高。

khadas@Khadas:~/ksnn/examples/tensorflow$ python3 inceptionv3.py --model ./models/VIM3/inceptionv3.nb --library ./libs/libnn_inceptionv3.so --picture ./data/goldfish_299x299.jpg --level 2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
 |--- KSNN Version: v1.3 +---| 
Start init neural network ...
#productname=VIPNano-QI, pid=0x88
Create Neural Network: 283ms or 283181us
Done.
Get input data ...
Done.
Start inference ...
Start run graph [1] times...
generate command buffer, total device count=1, core count per-device: 1, 
current device id=0, AXI SRAM base address=0xff000000
---------------------------Begin VerifyTiling -------------------------
AXI-SRAM = 1048576 Bytes VIP-SRAM = 522240 Bytes SWTILING_PHASE_FEATURES[1, 1, 0]
  0 NBG [(   0    0    0 0,        0, 0x(nil)(0x(nil), 0x(nil)) ->    0    0    0 0,        0, 0x(nil)(0x(nil), 0x(nil))) k(0 0    0,        0) pad(0 0) pool(0 0, 0 0)]

 id IN [ x  y  w   h ]   OUT  [ x  y  w  h ] (tx, ty, kpc) (ic, kc, kc/ks, ks/eks, kernel_type)
   0 NBG DD 0x(nil) [   0    0        0        0] -> DD 0x(nil) [   0    0        0        0] (  0,   0,   0) (       0,        0, 0.000000%, 0.000000%, NONE)

PreLoadWeightBiases = 1048576  100.000000%
---------------------------End VerifyTiling -------------------------
layer_id: 0 layer name:network_binary_graph operation[0]:unkown operation type target:unkown operation target.
uid: 0
abs_op_id: 0
execution time:             20552 us
[     1] TOTAL_READ_BANDWIDTH  (MByte): 67.540481
[     2] TOTAL_WRITE_BANDWIDTH (MByte): 18.245340
[     3] AXI_READ_BANDWIDTH  (MByte): 30.711348
[     4] AXI_WRITE_BANDWIDTH (MByte): 15.229973
[     5] DDR_READ_BANDWIDTH (MByte): 36.829133
[     6] DDR_WRITE_BANDWIDTH (MByte): 3.015367
[     7] GPUTOTALCYCLES: 94344921
[     8] GPUIDLECYCLES: 78109663
VPC_ELAPSETIME: 118090
*********
Run the 1 time: 118.00ms or 118636.00us
vxProcessGraph execution time:
Total   118.00ms or 118996.00us
Average 119.00ms or 118996.00us
Done. inference :  0.1422710418701172
----- Show Top5 +-----
     2: 0.93457
   795: 0.00328
   408: 0.00158
   974: 0.00148
   393: 0.00093

可以看到相关的所有信息。

摄像头Demo

  1. 目前支持摄像头的Demo有Yolo系列和OpenPose。以Yolov3为例，
khadas@Khadas:~$ cd ksnn/examples/darknet/
khadas@Khadas:~/ksnn/examples/darknet$ python3 hand-cap.py --model ./models/VIM3/hand.nb --library ./libs/libnn_hand.so --device X
  1. 目前支持RTSP的demo只有yolo系列。以Yolov3为例，
khadas@Khadas:~$ cd ksnn/examples/darknet/
khadas@Khadas:~/ksnn/examples/darknet$ python3 flask-yolov3.py --model ./models/VIM3/yolov3.nb --library ./libs/libnn_yolov3.so --device X

更多

KSNN转换工具使用说明
KSNN API文档