alex-petrenko/megaverse
High-throughput simulation platform for Artificial Intelligence reseach
This platform helps AI researchers efficiently train reinforcement learning agents in simulated 3D environments. It takes in configurations for complex virtual worlds and agent behaviors, then rapidly generates millions of visual observations and physics data. The primary users are AI and machine learning researchers developing and testing new algorithms for intelligent agents.
227 stars. No commits in the last 6 months.
Use this if you need to quickly simulate and render diverse 3D environments to train or evaluate AI agents, especially for tasks involving physics like building or navigating obstacles.
Not ideal if you're looking for a simple game engine for interactive applications or require extremely high-fidelity photorealistic rendering for visual effects.
Stars
227
Forks
21
Language
C++
License
MIT
Category
Last pushed
Dec 01, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/alex-petrenko/megaverse"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
facebookresearch/habitat-lab
A modular high-level library to train embodied AI agents across a variety of tasks and environments.
facebookresearch/ELF
An End-To-End, Lightweight and Flexible Platform for Game Research
utra-robosoccer/soccerbot
Soccer playing robot representing Canada from University of Toronto
Project-DC/pygeneses
A PyTorch based DeepRL framework to train and study artificial species in bio-inspired environments.
diambra/arena
DIAMBRA Arena: a New Reinforcement Learning Platform for Research and Experimentation