A 2D gridworld environment for reinforcement learning with single- and multi-agent support. Configurable grid size, walls, start and end positions, and rewards. Integrates with `relayrl_framework` and Burn tensors.
Optional GUI rendering (iced) is behind the `render` feature:
```toml
gridworld-rl = { version = "0.1.1", features = ["render"] }
```
[Apache License 2.0](../../LICENSE)