https://github.com/bonifatius94/rl-algos
SOTA Reinforcement Learning Algorithms
https://github.com/bonifatius94/rl-algos
artificial-intelligence dreamer-v2 ppo python reinforcement-learning tensorflow
Last synced: about 2 months ago
JSON representation
SOTA Reinforcement Learning Algorithms
- Host: GitHub
- URL: https://github.com/bonifatius94/rl-algos
- Owner: Bonifatius94
- License: mit
- Created: 2023-02-10T11:59:11.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-08-16T13:32:13.000Z (almost 3 years ago)
- Last Synced: 2025-02-25T06:18:35.791Z (over 1 year ago)
- Topics: artificial-intelligence, dreamer-v2, ppo, python, reinforcement-learning, tensorflow
- Language: Python
- Homepage:
- Size: 95.7 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Reference RL Algorithm Implementation
## About
This repository outlines several reinforcement learning algorithms
and uses them to solve common OpenAI Gym environment tasks.
The algorithm outlined is Proximal Policy Optimization (PPO). The implementation
of Dreamer is currently not working. Feel free to contribute in case you want
to help implementing more algorithms.
## Quickstart
```sh
git clone https://github.com/Bonifatius94/rl-algos
cd rl-algos
```
### Local Deployment
```sh
python3 -m pip install virtualenv
virtualenv venv --python=python3
source venv/bin/activate
pip install -r build_requirements.txt
pip install -r runtime_requirements.txt
deactivate
```
### Run Training With Visualization
```sh
python3 train_interactive.py
```
### Docker Deployment
```sh
docker-compose build && docker-compose run \
rlalgos-cuda python ./train_headless.py
```