Our great sponsors
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
If you want something as simple as possible, I'd go with MiniGrid, and if you want to have a richer world with more complex settings, then MiniHack.
If you want something as simple as possible, I'd go with MiniGrid, and if you want to have a richer world with more complex settings, then MiniHack.
GridVerse was specifically designed to allow users to create custom gridworld environments ( https://github.com/abaisero/gym-gridverse ) . Disclaimer: I'm the lead developer. We are still making some minor changes to the library, but the core of it is well founded, and we're getting close to a version 1.0.0, and I've already used it for published work. The documentation (for now you'll have to build it locally, the version on readthedocs is not up-to-date since we're in the middle of writing a comprehensive tutorial) contains a lot of examples on how to create your own custom environments, combining different transition, observation, reward functions, etc.
Thank you everyone! In the end, I created a new simple environment from scratch. If you’re interested you can check it out at https://github.com/damat-le/gym-simplegrid
Related posts
- Using FastAI to navigate matterport spaces?
- How to train an agent in custom mini-grid environment using stable baselines3?
- What OpenAI Gym environments are your favourite for learning RL algorithms?
- RL environment for hard exploration (infinite) task
- [R] Are there any paper about reinforcement learning solving mazes?