MineRL: Towards AI in Minecraft

Welcome to documentation for the MineRL project and its related repositories and components!

What is MineRL

MineRL is a rich Python 3 library which provides a OpenAI Gym interface for interacting with the video game Minecraft, accompanied with datasets of human gameplay.

Started as a research project at Carnegie Mellon University, MineRL aims to assist in the develpment of various aspects of artificial intelligence within Minecraft.

Tutorials and Guides

• Installation
• Hello World: Your First Agent
  • Creating an environment
  • Taking actions
• More Tutorials

MineRL Environments

• General Information
  • Observation Space
  • Action Space
• MineRL BASALT Competition Environments
  • MineRLBasaltFindCave-v0
  • MineRLBasaltCreateVillageAnimalPen-v0
  • MineRLBasaltMakeWaterfall-v0
  • MineRLBasaltBuildVillageHouse-v0
• MineRL Obtain Diamond Environments
  • MineRLObtainDiamondShovel-v0

Notes

• Performance tips
  • Faster alternative to xvfb
  • Docker images for headless rendering with GPU
  • Singularity container for headless rendering with GPU
• Links to papers and projects
  • Presentations
  • MineRL papers
  • 2019 competitor code/papers
  • 2020 competitor code/papers
  • Other papers that use the MineRL environment
  • Other
• General FAQ
  • For Version 1.x
  • For Version 0.4.x
• Windows FAQ
  • The The system cannot find the path specified error (installing)
  • The freeze_support error (multiprocessing)
• MineRL Versions
• Other Minecraft Interfaces
  • Project Malmo
  • MarLÖ “Multi-Agent Reinforcement Learning in MalmÖ”
  • MalmoEnv
  • IGLU “Interactive Grounded Language Understanding in a Collaborative Environment”
  • MineDojo

Indices and tables

• Index

• Module Index

• Search Page