DQN and Q Learning and Solving Environments

Hello, Community!

I am very new and I have a question that the web and documentation could not provide a direct answer.

I was intrigued with the visual aspect of the perceptilabs that is the very reason I started using the program.

My question is how am I able to construct a Deep Q Learning with custom gym environments in perceptilabs. Each time I click on create it want me to load a CSV file which is fine but my main concern is the data does not have a classification problem. My main goal is to solve an environment with data or create data. Maybe this isn’t the tool for me to use but I would like to learn how to use this tool to construct my ideas to solve an environment IF possible.

Thank you for your time!

Hi @FireKookies,
Welcome to the forum!

Unfortunately, Reinforcement Learning is not currently supported by the latest version of the tool. It’s something we will look at re-adding at some point, but we don’t have any ETA for it yet.
However, the older iteration of the tool (version 0.11.15) is still downloadable through pip and supports RL in case you want to give that a go :slight_smile:

All the best,
Robert