While going through SpinningUp as Deep RL Researcher, I found it challenging to connect and understand the paper (theory) and available implementation (practical) of several algorithms.

In this repository, I try to implement some of the most important, foundational Deep RL algorithms in a single notebook without any dependencies. There may be a lot of redundant code due to this.
You can directly use these notebooks in Google Colab.

Folder-wise readme contains the references and other resources used for that implementation.

You can read about my journey in RL here.
Here is the path I would suggest to a beginner in RL.