Harley Wiltzer
Harley Wiltzer
Firstly, I very much agree with your first step, so I'll get on that! > Different variants have really different parameters, so I'm afraid the definition of DQNLearner would be...
I like where this is going. Would it make sense to make something like `QNetworkWithTarget
This raised an alarm for me as well, but I checked and indeed the gymnax environments all do auto-reset. See https://github.com/RobertTLange/gymnax/blob/aef77d5c642ea48b95f34c51d05b8417d9450e15/gymnax/environments/environment.py#L48-L51.