Deepstream 6.0 Python Bindings Docs are terrible

Docs URL: https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/tree/v1.1.1/bindings

More than a week fighting to compile python bindings and put them into the docker container, between the x86, the Jetson and the Docker container, the steps don’t tell where to run each command, it’s frustrating… someone has some personal guide to complete this task?

1 Like

Can you refer to the FAQ?
DeepStream SDK FAQ - Intelligent Video Analytics / DeepStream SDK - NVIDIA Developer Forums

Hi @socieboy, thanks for the feedback and we are looking into improving the docs.

Meanwhile, we can guide you through the build process if you can share some info:

  1. Please note that we just released v1.1.2 along with DeepStream SDK 6.1. That is the recommended version. This release is based on Ubuntu 20.04. Would you be able to upgrade to this latest version?
  2. Why are you building the bindings yourself instead of using prebuilt wheels from our release section: Releases · NVIDIA-AI-IOT/deepstream_python_apps · GitHub
    Do you need customization in the bindings?
  3. Which platform are you building for, x86 or Jetson?

Thanks for your response, I was able to figure out what the process needs to be, i create the wheels but i found later that they were able to be downloaded from the sources page, this is what i mean with the documentation is lack and confusing, as a developer a understand that we fail on writing user guides because we asume that they know what we already know, and that’s not the case.
This topic can be close but I sugest to make the docs much detailed as possible.
Thanks

2 Likes

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.

We have updated documentation to make the bindings usage and build process more clear. Thanks again and your feedback is always appreciated.

1 Like