hill-a / stable-baselines

A fork of OpenAI Baselines, implementations of reinforcement learning algorithms
http://stable-baselines.readthedocs.io/
MIT License
4.14k stars 723 forks source link

VecNormalize 'training' attribute #1187

Open madhekar opened 1 year ago

madhekar commented 1 year ago

I have general question about VecNormalize 'training' attribute. I pre assume, I need to create my custom env instance for predict using VecNormalize? Should training=False? What is purpose of this attribute?

VecNormalize(env_learn, training=True, norm_obs=True, norm_reward=True)