An open API service indexing awesome lists of open source software.

https://github.com/robertoschiavone/flappy-bird-env

Flappy Bird as a Farama Gymnasium environment.
https://github.com/robertoschiavone/flappy-bird-env

flappy-bird gymnasium reinforcement-learning

Last synced: 24 days ago
JSON representation

Flappy Bird as a Farama Gymnasium environment.

Awesome Lists containing this project

README

          

# Flappy Bird Env


flappy bird



Action Space
Discrete(2)


Observation Shape
(800, 576, 3)


Observation High
255


Observation Low
0


Import
import flappy_bird_env # noqa
gymnasium.make("FlappyBird-v0")

## Description

Flappy Bird as a Farama Gymnasium environment.

### Installation

```bash
pip install flappy-bird-env
```

### Usage

1. Play it by running

```bash
python -m flappy_bird_env
```

Press `space` to flap the wings.

2. Import it to train your RL model

```python
import flappy_bird_env # noqa
env = gymnasium.make("FlappyBird-v0")
```

The package relies on ```import``` side-effects to register the environment
name so, even though the package is never explicitly used, its import is
necessary to access the environment.

## Action Space

Flappy Bird has the action space `Discrete(2)`.

| Value | Meaning |
|-------|------------|
| 0 | NOOP |
| 1 | flap wings |

## Observation Space

The observation will be the RGB image that is displayed to a human player with
observation space `Box(low=0, high=255, shape=(800, 576, 3), dtype=np.uint8)`.

### Rewards

You get `+1` every time you pass a pipe, otherwise `+0.001` for each frame where you
don't collide against the top and bottom bounds, or against a pipe.

## Version History

- v0: initial version release