Eclectic-Sheep/sheeprl
A distributed reinforcement learning library using Lightning Fabric for scalable agent training.
Not currently ranked — collecting fresh signals.
star history
SheepRL is a Python framework for training reinforcement learning agents at scale using distributed computing. It leverages Lightning Fabric to enable multi-GPU and multi-node training of RL algorithms, supporting environments like Atari games and Crafter. The library provides implementations of RL algorithms with benchmark results demonstrating competitive performance.
Frequently asked
- What is Eclectic-Sheep/sheeprl?
- A distributed reinforcement learning library using Lightning Fabric for scalable agent training.
- Is sheeprl open source?
- Yes — Eclectic-Sheep/sheeprl is open source, released under the Apache-2.0 license.
- What language is sheeprl written in?
- Eclectic-Sheep/sheeprl is primarily written in Python.
- How popular is sheeprl?
- Eclectic-Sheep/sheeprl has 436 stars on GitHub.
- Where can I find sheeprl?
- Eclectic-Sheep/sheeprl is on GitHub at https://github.com/Eclectic-Sheep/sheeprl.