This repository serves as a playground for experimenting with various Reinforcement Learning (RL) algorithms, including TRPO, PPO, DDPG, TD3, and SAC. Different algorithms have been implemented, and their performance, along with training times, have been measured for comparison. The project uses Stable Baselines3.
This project uses the poetry package manager for dependency management. Make sure you have poetry installed. If not, you can install it following the instructions here.
To set up this project, follow these steps:
-
Clone the repository:
git clone https://github.com/eisenmsi/RL-Playground.git cd RL-Playground -
Install the dependencies using poetry:
poetry install
Each RL algorithm is implemented in separate files (TRPO, PPO, DDPG, TD3, SAC). Navigate to the specific file to explore the implementation details and configuration options.
To train and evaluate an algorithm, run the respective file.
Performance metrics and training times for each algorithm are determined by running the files.
This project is licensed under the MIT License.