hill-a / stable-baselines

A fork of OpenAI Baselines, implementations of reinforcement learning algorithms
http://stable-baselines.readthedocs.io/
MIT License
4.16k stars 725 forks source link

Can I use an agent with act, and observe interactions with no/minimum use of environment? #1178

Open aheidariiiiii1993 opened 1 year ago

aheidariiiiii1993 commented 1 year ago

Hello, I am trying to use an agent for a control problem, in which my agent will be in Python as a component of a simulation software. The agent needs to interact with the simulated energy system (perform actions, and receive back next state and reward values) at each timestep of simulation. Would you please let me know if it is straightforward in stable-baselines library? I have been using another library so far, but since their user forum is not working anymore I have to imigrate to another library. If so,, would you please share an exampleif exists.

Thanks