Jetson Nano : Jetson Inference

출처

• Getting started with the NVIDIA Jetson Nano - PyImageSearch
• Is there any demos available for python jetson inference - NVIDIA Developer Forums
• Official TensorFlow for Jetson Nano !!! - NVIDIA Developer Forums
• Install OpenCV 4 on your Raspberry Pi - PyImageSearch
• [OpenCV 설치] Ubuntu OpenCV 설치, C++ 테스트
• How to find OpenCV version in Python and C++ ? | Learn OpenCV
• 4.0.0 does not produce pkgconfig file · Issue #13154 · opencv/opencv · GitHub

라즈비안(Raspberry pi OS)의 업데이트와 업그레이드

$ sudo apt-get update && sudo apt-get upgrade

OpenCV 4.1 관련 패키지 설치

OpenCV 빌드 관련 도구 설치

$ sudo apt-get install git cmake
$ sudo apt-get install libatlas-base-dev gfortran
$ sudo apt-get install libhdf5-serial-dev hdf5-tools
$ sudo apt-get install python3-dev

OpenCV 4.1 용 Python 3 가상 환경 구성

pip 설치

$ wget https://bootstrap.pypa.io/get-pip.py
$ sudo python3 get-pip.py
WARNING: The directory '/home/bluesanta/.cache/pip/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
WARNING: The directory '/home/bluesanta/.cache/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
Collecting pip
  Downloading https://files.pythonhosted.org/packages/62/ca/94d32a6516ed197a491d17d46595ce58a83cbb2fca280414e57cd86b84dc/pip-19.2.1-py2.py3-none-any.whl (1.4MB)
     |████████████████████████████████| 1.4MB 150kB/s 
Collecting setuptools
  Downloading https://files.pythonhosted.org/packages/ec/51/f45cea425fd5cb0b0380f5b0f048ebc1da5b417e48d304838c02d6288a1e/setuptools-41.0.1-py2.py3-none-any.whl (575kB)
     |████████████████████████████████| 583kB 379kB/s 
Collecting wheel
  Downloading https://files.pythonhosted.org/packages/bb/10/44230dd6bf3563b8f227dbf344c908d412ad2ff48066476672f3a72e174e/wheel-0.33.4-py2.py3-none-any.whl
ERROR: launchpadlib 1.10.6 requires testresources, which is not installed.
Installing collected packages: pip, setuptools, wheel
Successfully installed pip-19.2.1 setuptools-41.0.1 wheel-0.33.4
$ rm get-pip.py

virtualenv, virtualenvwrapper 설치

$ sudo pip install virtualenv virtualenvwrapper

python3, virtualenv 환경설정

$ echo -e "\n# virtualenv and virtualenvwrapper" >> ~/.profile
$ echo "export WORKON_HOME=$HOME/.virtualenvs" >> ~/.profile
$ echo "export VIRTUALENVWRAPPER_PYTHON=/usr/bin/python3" >> ~/.profile
$ echo "source /usr/local/bin/virtualenvwrapper.sh" >> ~/.profile
$ source ~/.profile

가상환경을 만들기

$ mkvirtualenv cv -p python3
Already using interpreter /usr/bin/python3
Using base prefix '/usr'
New python executable in /home/pi/.virtualenvs/cv/bin/python3
Also creating executable in /home/pi/.virtualenvs/cv/bin/python
Installing setuptools, pip, wheel...
done.
virtualenvwrapper.user_scripts creating /home/pi/.virtualenvs/cv/bin/predeactivate
virtualenvwrapper.user_scripts creating /home/pi/.virtualenvs/cv/bin/postdeactivate
virtualenvwrapper.user_scripts creating /home/pi/.virtualenvs/cv/bin/preactivate
virtualenvwrapper.user_scripts creating /home/pi/.virtualenvs/cv/bin/postactivate
virtualenvwrapper.user_scripts creating /home/pi/.virtualenvs/cv/bin/get_env_details

workon 명령을 사용하여 cv 환경에 있는지 확인

pi@raspberrypi:~$ workon cv
(cv) pi@raspberrypi:~$ 

numpy 파이썬 패키지 설치(OpenCV관련 수학 함수 모음)

pi@raspberrypi:~$ workon cv
(cv) pi@raspberrypi:~$ pip install numpy
Looking in indexes: https://pypi.org/simple, https://www.piwheels.org/simple
Collecting numpy
  Using cached https://www.piwheels.org/simple/numpy/numpy-1.16.4-cp37-cp37m-linux_armv7l.whl
Installing collected packages: numpy
Successfully installed numpy-1.16.4
(cv) pi@raspberrypi:~$ python
Python 3.7.3 (default, Apr  3 2019, 05:39:12) 
[GCC 8.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy
>>> quit()

TensorFlow 설치 : NVIDIA는 Jetson Nano의 TensorFlow 공식 릴리스를 제공

$ pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v42 tensorflow-gpu==1.13.1+nv19.3
 
Successfully installed absl-py-0.7.1 astor-0.8.0 gast-0.2.2 grpcio-1.22.0 h5py-2.9.0 keras-applications-1.0.8 keras-preprocessing-1.1.0 markdown-3.1.1 mock-3.0.5 protobuf-3.9.0 six-1.12.0 tensorboard-1.13.1 tensorflow-estimator-1.13.0 tensorflow-gpu-1.13.1+nv19.3 termcolor-1.1.0 werkzeug-0.15.5

scipy, keras 설치

$ pip install scipy
$ pip install keras

Jetson Inference 컴파일 및 설치

$ cd ~
$ git clone https://github.com/dusty-nv/jetson-inference
$ cd jetson-inference
$ git submodule update --init
$ mkdir build
$ cd build/
$ cmake ../
$ make
$ sudo make install

이미지 인식 예제 실행

$ cd aarch64/bin
$ python3 my-recognition.py --network=googlenet black_bear.jpg
jetson.inference.__init__.py
jetson.inference -- initializing Python 3.6 bindings...
jetson.inference -- registering module types...
jetson.inference -- done registering module types
jetson.inference -- done Python 3.6 binding initialization
jetson.utils.__init__.py
jetson.utils -- initializing Python 3.6 bindings...
jetson.utils -- registering module functions...
jetson.utils -- done registering module functions
jetson.utils -- registering module types...
jetson.utils -- done registering module types
jetson.utils -- done Python 3.6 binding initialization
[image] loaded 'black_bear.jpg'  (800 x 656, 3 channels)
jetson.inference -- PyTensorNet_New()
jetson.inference -- PyImageNet_Init()
jetson.inference -- imageNet loading build-in network 'googlenet'

imageNet -- loading classification network model from:
         -- prototxt     networks/googlenet.prototxt
         -- model        networks/bvlc_googlenet.caffemodel
         -- class_labels networks/ilsvrc12_synset_words.txt
         -- input_blob   'data'
         -- output_blob  'prob'
         -- batch_size   1

[TRT]   TensorRT version 5.1.6
[TRT]   loading NVIDIA plugins...
[TRT]   Plugin Creator registration succeeded - GridAnchor_TRT
[TRT]   Plugin Creator registration succeeded - NMS_TRT
[TRT]   Plugin Creator registration succeeded - Reorg_TRT
[TRT]   Plugin Creator registration succeeded - Region_TRT
[TRT]   Plugin Creator registration succeeded - Clip_TRT
[TRT]   Plugin Creator registration succeeded - LReLU_TRT
[TRT]   Plugin Creator registration succeeded - PriorBox_TRT
[TRT]   Plugin Creator registration succeeded - Normalize_TRT
[TRT]   Plugin Creator registration succeeded - RPROI_TRT
[TRT]   Plugin Creator registration succeeded - BatchedNMS_TRT
[TRT]   completed loading NVIDIA plugins.
[TRT]   detected model format - caffe  (extension '.caffemodel')
[TRT]   desired precision specified for GPU: FASTEST
[TRT]   requested fasted precision for device GPU without providing valid calibrator, disabling INT8
[TRT]   native precisions detected for GPU:  FP32, FP16
[TRT]   selecting fastest native precision for GPU:  FP16
[TRT]   attempting to open engine cache file networks/bvlc_googlenet.caffemodel.1.1.GPU.FP16.engine
[TRT]   cache file not found, profiling network model on device GPU
[TRT]   device GPU, loading networks/googlenet.prototxt networks/bvlc_googlenet.caffemodel
[TRT]   retrieved Output tensor "prob":  1000x1x1
[TRT]   retrieved Input tensor "data":  3x224x224
[TRT]   device GPU, configuring CUDA engine
[TRT]   device GPU, building FP16:  ON
[TRT]   device GPU, building INT8:  OFF
[TRT]   device GPU, building CUDA engine (this may take a few minutes the first time a network is loaded)
[TRT]   device GPU, completed building CUDA engine
[TRT]   network profiling complete, writing engine cache to networks/bvlc_googlenet.caffemodel.1.1.GPU.FP16.engine
[TRT]   device GPU, completed writing engine cache to networks/bvlc_googlenet.caffemodel.1.1.GPU.FP16.engine
[TRT]   device GPU, networks/bvlc_googlenet.caffemodel loaded
[TRT]   device GPU, CUDA engine context initialized with 2 bindings
[TRT]   binding -- index   0
               -- name    'data'
               -- type    FP32
               -- in/out  INPUT
               -- # dims  3
               -- dim #0  3 (CHANNEL)
               -- dim #1  224 (SPATIAL)
               -- dim #2  224 (SPATIAL)
[TRT]   binding -- index   1
               -- name    'prob'
               -- type    FP32
               -- in/out  OUTPUT
               -- # dims  3
               -- dim #0  1000 (CHANNEL)
               -- dim #1  1 (SPATIAL)
               -- dim #2  1 (SPATIAL)
[TRT]   binding to input 0 data  binding index:  0
[TRT]   binding to input 0 data  dims (b=1 c=3 h=224 w=224) size=602112
[TRT]   binding to output 0 prob  binding index:  1
[TRT]   binding to output 0 prob  dims (b=1 c=1000 h=1 w=1) size=4000
device GPU, networks/bvlc_googlenet.caffemodel initialized.
[TRT]   networks/bvlc_googlenet.caffemodel loaded
imageNet -- loaded 1000 class info entries
networks/bvlc_googlenet.caffemodel initialized.
class 0295 - 0.989919  (American black bear, black bear, Ursus americanus, Euarctos americanus)
image is recognized as 'American black bear, black bear, Ursus americanus, Euarctos americanus' (class #295) with 98.991907% confidence
jetson.utils -- freeing CUDA mapped memory
PyTensorNet_Dealloc()

예제 : 카메라를 사용하여 이미지 분류

컴파일

$ cd ~/jetson-inference/build/
$ make imagenet-camera
[100%] Built target imagenet-camera

실행