https://github.com/franco-ruggeri/kth-reinforcement-learning
Assignments of the Reinforcement Learning course (EL2805) at KTH
https://github.com/franco-ruggeri/kth-reinforcement-learning
reinforcement-learning
Last synced: 8 months ago
JSON representation
Assignments of the Reinforcement Learning course (EL2805) at KTH
- Host: GitHub
- URL: https://github.com/franco-ruggeri/kth-reinforcement-learning
- Owner: franco-ruggeri
- License: mit
- Created: 2022-11-07T22:23:52.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-01-27T16:56:52.000Z (over 2 years ago)
- Last Synced: 2025-01-08T08:46:14.450Z (9 months ago)
- Topics: reinforcement-learning
- Language: Python
- Homepage:
- Size: 6.73 MB
- Stars: 0
- Watchers: 4
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Reinforcement Learning
Assignments of the Reinforcement Learning course (EL2805) at KTH.
The repository provides a [Python package](el2805), which includes:
- [Dynamic programming algorithms for the exact solution of MDPs](el2805/agents/mdp)
- [RL algorithms](el2805/agents/rl)
- [Simulators for problems modeled as MDPs](el2805/envs)
You can install the Python package as follows:
```shell
pip install -e .
```
Additionally, the repository contains [documentation](docs) (instructions and reports), [scripts](scripts)
and [tests](tests). You can run scripts and tests as follows:
```shell
python scripts/ # run a particular script
python -m unittest # run all tests
```
Available algorithms:
- [x] [Dynamic programming](el2805/agents/mdp/dynamic_programming.py)
- [x] [Value iteration](el2805/agents/mdp/value_iteration.py)
- [x] [Q-learning](el2805/agents/rl/q_learning.py)
- [x] [SARSA](el2805/agents/rl/sarsa.py)
- [ ] λ-SARSA
- [x] [DQN](el2805/agents/rl/dqn.py)
- [ ] DDPG
- [x] [PPO](el2805/agents/rl/ppo.py)
The missing algorithms have been implemented by my lab partner and can be found
in [his repository](https://github.com/afcarzero1/ReinforcementLearning).