pop-jetson 0.0.6

AIoT AI library for pop

How to use

Install dependencies

    sudo apt install libjpeg-dev zlib1g-dev libopenblas-dev libavcodec-dev libavformat-dev libswscale-dev

Install onnx-graphsurgeon from NVidia pre-build source files.

    pip install https://developer.download.nvidia.com/compute/redist/onnx-graphsurgeon/onnx_graphsurgeon-0.3.27-py2.py3-none-any.whl

Install PyTorch from NVidia pre-built source files.

    #For python 3.8 
    pip install https://developer.download.nvidia.cn/compute/redist/jp/v511/pytorch/torch-2.0.0+nv23.05-cp38-cp38-linux_aarch64.whl

    #For python 3.6 
    pip install https://developer.download.nvidia.cn/compute/redist/jp/v461/pytorch/torch-1.11.0a0+17540c5+nv22.01-cp36-cp36m-linux_aarch64.whl

Once you have done installing PyTorch, install TorchVision with following commands

    #For python 3.8 
    export BUILD_VERSION=0.15.2
    git clone --branch v0.15.2 https://github.com/pytorch/vision torchvision   

    #For python 3.6
    export BUILD_VERSION=0.12.0
    git clone --branch v0.12.0 https://github.com/pytorch/vision torchvision 

When the clonning is over, type:

    cd torchvision
    python3 setup.py install --user
    cd ..

    # To install it on system
    cd torchvision
    sudo python3 setup.py build
    sudo pip install .
    cd ..