Metadata-Version: 2.1
Name: mazenv
Version: 0.4.2
Summary: Maze environments for Reinforcement Learning
Home-page: https://github.com/unixpickle/mazenv-py
Author: Alex Nichol
Author-email: unixpickle@gmail.com
License: BSD
Keywords: ai reinforcement learning
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: BSD License
Classifier: Programming Language :: Python :: 3.5
Classifier: Programming Language :: Python :: 3.6
Requires-Dist: numpy (<2.0.0,>=1.0.0)
Requires-Dist: gym (<0.11.0,>=0.10.0)

# mazenv-py

A Python port of [unixpickle/mazenv](https://github.com/unixpickle/mazenv) that uses OpenAI Gym.

# Installation

You can install mazenv with pip:

```
pip install mazenv
```

# Usage

You can generate a random 8x8 maze like this:

```
import mazenv

maze = mazenv.prim((8, 8))
print(maze)
```

You can create a Gym environment out of your maze like this:

```
env = mazenv.Env(maze)
```

If you want to restrict the observations to a 5x5 grid centered around the current position, you can do:

```
restricted_env = mazenv.HorizonLimit(env, horizon=2)
```


