TD 1 : Hands-On Reinforcement Learning

MSO 3.4 Apprentissage Automatique

Installation

Modules to be installed in order to start the project :

pip install gym==0.26.2
pip install pyglet==2.0.10
pip install pygame==2.5.2
pip install PyQt5

pip install stable-baselines3
pip install moviepy

REINFORCE

You will find the reinforcement algorithm on the .py file : 'reinforce_cartpole.py'.

Here is the final plot obtained after running it, showing the total reward after the 200 episodes.

Stable-Baselines3 and HuggingFace

We are now solving the CartPole with the A2C algorithm, from the Stable-Baselines3 package. We saved the model through Hugging Face, which is available via the following link : https://huggingface.co/jossuaseksik/a2c-sb3_cartpole/tree/main