Clarification on In-Context Reinforcement Learning Implementation #2

Hello,

First of all, great project—thank you for all your hard work!

However, I have been reviewing the main.py file and noticed that there seems to be no implementation of in-context reinforcement learning (ICRL). While I found a calc_reward function, it appears unused. Additionally, there seems to be no implementation of a neural network, which I expected might be part of an ICRL setup.

Could you please explain how in-context RL is applied in your code?

Thank you!