jetson yolo environment_python3.6编译torch==1.9.0下载-CSDN博客

本文链接：https://blog.csdn.net/longwoo1012/article/details/123831144

Tool

Host Computer Ubuntu18
download SDKManager from nvidia
usb connected
Jetson and Host connected in the same router
Enter recorver mode (connect ground and REC pin 2, pin3)
Power ON (J48 need connect)

OS

Jetpack 4.6.1
Develop Tools(Cuda , cudnn …)
Select max Power Mode after OS installed
Start fan sudo sh -c 'echo 255 >/sys/devices/pwm-fan/target_pwm'

Yolo Dependence

first go to https://forums.developer.nvidia.com/t/pytorch-for-jetson-version-1-10-0-now-available/72048 and download pytorch1.9.0 file torch-1.9.0-cp36-cp36m-linux_aarch64.whl
download torchvision0.10.0 git clone --branch 0.10.0 https://github.com/pytorch/vision torchvision
Install source dependence

sudo apt-get install libblas3 liblapack3 liblapack-dev libblas-dev gfortran libatlas-base-dev
sudo apt-get install libjpeg-dev zlib1g-dev libpython3-dev libavcodec-dev libavformat-dev libswscale-dev
sudo apt-get install libfreetype6-dev
sudo apt-get install python3-pip libopenblas-base libopenmpi-dev 
pip3 install setuptools -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install Cython -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install numpy -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install Pillow -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install sklearn -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install serial  -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install networkx -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install tqdm -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install seaborn -i https://pypi.tuna.tsinghua.edu.cn/simple

Install pytorch1.9.0 and torchvision0.10.0 from source

pip3 install torch-1.9.0-cp36-cp36m-linux_aarch64.whl
$ cd torchvision
$ export BUILD_VERSION=0.10.0  # where 0.x.0 is the torchvision version  
$ python3 setup.py install --user
$ cd ../  # attempting to load torchvision from build dir will result in import error
$ #pip3 install 'pillow<7' # always needed for Python 2.7, not needed torchvision v0.5.0+ with Python

Yolo inference environment

disable all thop by comment

build opencv-4.4.0 and contrib-4.4.0 from source

https://automaticaddison.com/how-to-install-opencv-4-5-on-nvidia-jetson-nano/
https://github.com/opencv/opencv_contrib/issues/1301
https://blog.csdn.net/u011736771/article/details/85960300

sudo apt-get purge libopencv*
sudo apt-get install build-essential cmake git unzip pkg-config
sudo apt-get install libjpeg-dev libpng-dev libtiff-dev
sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev
sudo apt-get install libgtk2.0-dev libcanberra-gtk*
sudo apt-get install python3-dev python3-numpy python3-pip
sudo apt-get install libxvidcore-dev libx264-dev libgtk-3-dev
sudo apt-get install libtbb2 libtbb-dev libdc1394-22-dev
sudo apt-get install libv4l-dev v4l-utils
sudo apt-get install libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev
sudo apt-get install libavresample-dev libvorbis-dev libxine2-dev
sudo apt-get install libfaac-dev libmp3lame-dev libtheora-dev
sudo apt-get install libopencore-amrnb-dev libopencore-amrwb-dev
sudo apt-get install libopenblas-dev libatlas-base-dev libblas-dev
sudo apt-get install liblapack-dev libeigen3-dev gfortran
sudo apt-get install libhdf5-dev protobuf-compiler
sudo apt-get install libprotobuf-dev libgoogle-glog-dev libgflags-dev
cd ~
wget -O opencv.zip https://github.com/opencv/opencv/archive/4.4.0.zip
wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/4.4.0.zip
unzip opencv.zip
unzip opencv_contrib.zip
mv opencv-4.4.0 opencv
mv opencv_contrib-4.4.0 opencv_contrib
rm opencv.zip
rm opencv_contrib.zip
cd ~/opencv
mkdir build
cd build

cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr \
-D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules \
-D EIGEN_INCLUDE_PATH=/usr/include/eigen3 \
-D WITH_OPENCL=OFF \
-D WITH_CUDA=ON \
-D CUDA_ARCH_BIN=5.3 \
-D CUDA_ARCH_PTX="" \
-D WITH_CUDNN=ON \
-D WITH_CUBLAS=ON \
-D ENABLE_FAST_MATH=ON \
-D CUDA_FAST_MATH=ON \
-D OPENCV_DNN_CUDA=ON \
-D ENABLE_NEON=ON \
-D WITH_QT=OFF \
-D WITH_OPENMP=ON \
-D WITH_OPENGL=ON \
-D BUILD_TIFF=ON \
-D WITH_FFMPEG=ON \
-D WITH_GSTREAMER=ON \
-D WITH_TBB=ON \
-D BUILD_TBB=ON \
-D BUILD_TESTS=OFF \
-D WITH_EIGEN=ON \
-D WITH_V4L=ON \
-D WITH_LIBV4L=ON \
-D OPENCV_ENABLE_NONFREE=ON \
-D INSTALL_C_EXAMPLES=OFF \
-D INSTALL_PYTHON_EXAMPLES=OFF \
-D BUILD_NEW_PYTHON_SUPPORT=ON \
-D BUILD_opencv_python3=TRUE \
-D OPENCV_GENERATE_PKGCONFIG=ON \
-D BUILD_EXAMPLES=OFF ..

make -j4
cd ~/opencv/build
sudo make install
sudo ldconfig
make clean
sudo apt-get update

使用opencv 4.4.0的原因在于，从这一版开始，专利过期的sift算子从contribute模块的xfeature2d转移到主代码了，更方便调用。

docker experiment

just for fun

sudo docker pull nvcr.io/nvidia/l4t-ml:r32.6.1-py3
sudo docker run -it --gpus all -v /home/nano/yolov5-face-master/:/yolov5 -e DISPLAY=:0 -v /tmp/.X11-unix:/tmp/.X11-unix nvcr.io/nvidia/l4t-ml:r32.6.1-py3