NVIDIA-AI-IOT / jetracer

An autonomous AI racecar using NVIDIA Jetson Nano
MIT License
1.06k stars 319 forks source link

workaround for installation jetracer on Jetson nx, Jetpack 4.4 #89

Closed Jiroh closed 2 years ago

Jiroh commented 3 years ago

1) Make microSD card from JetPack SDK 4.4. jetson-nx-developer-kit-sd-card-image-4-4.zip passward must be "jetson" 2) JetCard $ git clone https://github.com/NVIDIA-AI-IOT/jetcard.git $ cd jetcard nano $ git checkout jetpack_4.4 Edit "install.sh" sudo -H pip3 install -U numpy grpcio absl-py py-cpuinfo psutil portpicker six mock requests gast h5py==2.10.0 astor termcolor protobuf keras-applicat$ sudo -H pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v44 tensorflow install some dependensy before install $ sudo apt-get install libswscale-dev $ sudo apt-get install libavformat-dev then $ ./install.sh 3) torch2trt $ git clone https://github.com/NVIDIA-AI-IOT/torch2trt.git $ cd torch2trt $ sudo python3 setup.py install 4) Jetcam $ git clone https://github.com/NVIDIA-AI-IOT/jetcam.git $ cd jetcam $ sudo python3 setup.py install 5) JetRacer $ git clone https://github.com/NVIDIA-AI-IOT/jetracer.git $ cd jetracer $ sudo python3 setup.py install 6) Update jupyter $ sudo rm -rf jupyter_clickabe_image_widget $ git clone https://github.com/jaybdub/jupyter_clickable_image_widget $ cd jupyter_clickable_image_widget $ sudo pip3 install -e . $ jupyter labextension install js 7) may be required to modify install.py for jetcard sudo -H python3 -m pip install git+https://github.com/ipython/traitlets@4.3.x