https://github.com/neoneye/simon-arc-env
Abstraction and Reasoning Corpus as a Farama Gymnasium environment.
https://github.com/neoneye/simon-arc-env
arc-agi gym-environment gymnasium reinforcement-learning reinforcement-learning-environments
Last synced: about 2 months ago
JSON representation
Abstraction and Reasoning Corpus as a Farama Gymnasium environment.
- Host: GitHub
- URL: https://github.com/neoneye/simon-arc-env
- Owner: neoneye
- License: mit
- Created: 2023-12-17T14:54:54.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-01-12T14:35:47.000Z (over 1 year ago)
- Last Synced: 2025-04-05T03:41:15.690Z (6 months ago)
- Topics: arc-agi, gym-environment, gymnasium, reinforcement-learning, reinforcement-learning-environments
- Language: Python
- Homepage:
- Size: 703 KB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Simon ARC Env
Action Space
Discrete(10)
Observation Shape
(32, 32)
Observation High
255
Observation Low
0
Import
import simon_arc_env
gymnasium.make("SimonARC-v0")
## Description
[Abstraction and Reasoning Corpus](https://github.com/fchollet/ARC) as a [Farama Gymnasium](https://github.com/Farama-Foundation/Gymnasium) environment.
Related project: [ARCLE - ARC Learning Environment](https://github.com/ConfeitoHS/arcle).
### Installation
```bash
pip install simon-arc-env
```
### Usage
1. Play it by running
```bash
python examples/play.py
```
Press `space` to draw a pixel.
2. Import it to train your RL model
```python
import simon_arc_env
env = gymnasium.make("SimonARC-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
`SimonARC` has the action space `Discrete(10)`.
| Value | Key | Meaning |
|-------|-------------|-----------------------|
| 0 | TAB | Next page |
| 1 | Return | Submit drawing |
| 2 | Space | Set pixel |
| 3 | Arrow up | Move `cursor_y` up |
| 4 | Arrow down | Move `cursor_y` down |
| 5 | Arrow left | Move `cursor_x` left |
| 6 | Arrow right | Move `cursor_x` right |
| 7 | m | Increment color |
| 8 | n | Decrement color |
| 9 | s | Adjust canvas size |
| N/A | Escape | Exit game |
| N/A | i | Print info to console |
## Observation Space
Observation space `Box(low=0, high=255, shape=(32, 32), dtype=np.uint8)`.
### Rewards
If you solve the task you get `10000` in reward.
If you predict the `width` correct you get `100` in reward.
If you predict the `height` correct you get `100` in reward.
If your prediction have the correct histogram you get `100` in reward.
## Run tests
```bash
make test
```
## Version History
- v0: initial version release