rlpyt
A modular, high-throughput PyTorch framework for deep reinforcement learning research, supporting policy gradient, deep Q-learning, and Q-function policy gradient algorithms.
What is rlpyt?
rlpyt is a deep reinforcement learning framework implemented in PyTorch, providing modular and optimized versions of common RL algorithms like A2C, PPO, DQN, DDPG, and SAC. It solves the need for a high-throughput, research-oriented codebase that unifies infrastructure for policy gradient, deep Q-learning, and Q-function policy gradient methods.
Researchers and developers working on small- to medium-scale deep reinforcement learning projects who need a flexible, efficient framework for experimenting with and modifying RL algorithms.
Developers choose rlpyt for its modular design, support for parallel and multi-GPU execution, and comprehensive algorithm implementations, which accelerate RL research without the overhead of large-scale systems like OpenAI's Dota setup.
Overview
Reinforcement Learning in PyTorch
Use Cases
Best For
- Implementing and comparing policy gradient algorithms like A2C and PPO
- Experimenting with deep Q-learning variants including Double DQN and Rainbow components
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
