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!

Question Save brain based on Std. of Reward

Discussion in 'ML-Agents' started by Husishow, Apr 30, 2023.

  1. Husishow

    Husishow

    Joined:
    May 14, 2019
    Posts:
    1
    I'm trying to apply a previous NN to a problem that is scaled up. I keep seeing that at certain runs the agent behaves as expected, but sometimes it either gets too much or too little reward. I'd like to see what the agent is doing during those runs so I wanted to know if there is a way to make a checkpoint when the std is too large or too little?
     
    Husishow, Apr 30, 2023
    #1