arcverse is a library for creating synthesizing arc-agi like datasets with knobs to control different degrees of freedom.
Project description
Arcverse
For now, install from source:
git clone
cd arcverse
pip install -e .
Then, open a notebook and run:
from arcverse import ArcWorld
from arcverse.utils import render
world = ArcWorld(max_rows=10, max_cols=10)
This will create a world that is initialized with a set of random transforms.
print(world.transforms)
Then you can draw samples from this world as a bunch of starting grids and their corresponding transformed grids.
for sample in world.sample(5):
display(render(sample[0]))
display(render(sample[1]))
print("---")
For now, a static arc-like puzzle is created and can be accessed via:
world.get_puzzle()
Since object shapes are random, there is a good chance all five train examples might not converge to an obvious solution but this will be refined over time.
Example
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file arcverse-0.0.1a2.tar.gz.
File metadata
- Download URL: arcverse-0.0.1a2.tar.gz
- Upload date:
- Size: 13.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/2.0.1 CPython/3.13.1 Linux/6.8.0-53-generic
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
cb320c86f82145410424893ea10f44a71522466a14785632a0e360cb5e1add30
|
|
| MD5 |
3cf828cfaec27eabcd297a3195a2f062
|
|
| BLAKE2b-256 |
ef8230f1cdf444d1079ca08fb67c3133ae5f38a21d6fa31bf77e8b2f3b99f1f7
|
File details
Details for the file arcverse-0.0.1a2-py3-none-any.whl.
File metadata
- Download URL: arcverse-0.0.1a2-py3-none-any.whl
- Upload date:
- Size: 13.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/2.0.1 CPython/3.13.1 Linux/6.8.0-53-generic
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
7e4b1f3518818f2fb9603f6c21fa6b59d0b207046799c7b846c018221fb4d3e7
|
|
| MD5 |
39d07f10ae25f80b65f1328ed6e8fe0f
|
|
| BLAKE2b-256 |
5690f8c7c845dc1b0943c2912e7481cc563b0627c8b304ba43297d7d99a9d836
|