Search Unity

  1. Unity support for visionOS is now available. Learn more in our blog post.
    Dismiss Notice

Question Unity ML-Agents 1.0.2: Training stops almost immediately

Discussion in 'ML-Agents' started by k1tsunee2, Nov 25, 2022.

  1. k1tsunee2

    k1tsunee2

    Joined:
    Apr 25, 2020
    Posts:
    1
    Hey!
    So, I'm trying to train a Dynamic Difficulty Adjustment using Reinforcement Learning. I created my Agent script, the yaml file, everything. However, when I try to train my agent, it stops the training almost immediately.

    Here is my DdaAgent.cs script -> DdaAgent.cs (github.com) and my training yaml -> trainer_config.yaml (github.com)

    Did I do anything wrong? If you need more code, I'll provide, just ask!

    I'm using ML-Agents 1.0.2 with Unity 2020.3.25f1