Hello @dilip.patel
Please follow the instructions below and let me know how it goes.
- Run the container using following command:
sudo docker run -it --rm --net=host --runtime nvidia -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --device /dev/video0 --privileged nvcr.io/nvidia/deepstream:6.0-triton
-
After you are in the container, please run the script docker_python_setup.sh, it doesn’t require any other packages and should be able to setup python3.6 successfully in your container.
-
Next, please clone the deepstream python repo at /opt/nvidia/deepstream/deepstream-6.0/sources/ and install the following required packages:
apt install python3-gi python3-gst-1.0 python-gi-dev git python3 python3-pip cmake g++ build-essential libglib2.0-dev python3-dev python3.6-dev libglib2.0-dev-bin python-gi-dev libtool m4 autoconf automake
Note: this installs gst-python so you don’t need to build it separately before building the bindings
-
Follow the build instructions here: https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/tree/master/bindings#24—deepstream-sdk
-
Make sure your default pip points to python3.6 and install the wheel you built in the above step using pip install ./pyds-1.1.0-py3-none-linux_x86_64.whl
Note: If you see any errors in the installation which look like this: pip install ./pyds-1.1.0-py3-none-linux_x86_64.whl):
/usr/bin/python3: can’t find ‘main’ module in ‘/usr/share/python-wheels/pep517-0.8.2-py2.py3-none-any.whl/pep517/_in_process.py’ ,
Please try installing with the command below instead:
pip install --no-use-pep517 ./pyds-1.1.0-py3-none-linux_x86_64.whl
Please let me know if you face any issues here.
Thanks