Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.
  2. Unite 2023 Registrations are now LIVE! We are thrilled to announce that Unite 2023 Registration is now live and open for all!
    Dismiss Notice
New Forum User Notice Unite 2023 Registrations are now LIVE!

Receive images from camera sensor in python and use custom neural network to take action

Discussion in 'ML-Agents' started by harshagg02, Sep 7, 2021.

  1. harshagg02

    harshagg02

    Joined:
    Sep 6, 2021
    Posts:
    2
    Hi, I am new to Unity. I have a camera sensor on my agent and I want to receive the images on every time step in python and use my neural network in Pytorch to take decision. But, how do I receive those observation images?
    Thanks a lot
     
    harshagg02, Sep 7, 2021
    #1

  2. ruoping_unity

    ruoping_unity

    Unity Technologies

    Joined:
    Jul 10, 2020
    Posts:
    134
    To build your own custom networks, you can use our low-level python API. The DecisionSteps and TerminalSteps you get from each env.step() will contain all the observations from the sensors on your agent.
     
    ruoping_unity, Sep 8, 2021
    #2