https://github.com/ml4a/ml4a

A python library and collection of notebooks for making art with machine learning.
https://github.com/ml4a/ml4a

Last synced: 28 days ago
JSON representation

A python library and collection of notebooks for making art with machine learning.

• Host: GitHub
• URL: https://github.com/ml4a/ml4a
• Owner: ml4a
• License: mit
• Created: 2016-06-01T03:33:14.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2023-04-17T06:32:55.000Z (over 1 year ago)
• Last Synced: 2024-04-14T00:59:41.880Z (7 months ago)
• Language: Python
• Homepage: https://ml4a.net
• Size: 239 MB
• Stars: 1,568
• Watchers: 76
• Forks: 335
• Open Issues: 14
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


  


  

  


  Machine Learning for Artists

  






     

     

    

    



[ml4a](https://ml4a.net) is a Python library for making art with machine learning. It features:

* an API wrapping popular deep learning models with creative applications, including [StyleGAN2](https://github.com/NVLabs/stylegan2/), [SPADE](https://github.com/NVlabs/SPADE), [Neural Style Transfer](https://github.com/genekogan/neural_style), [DeepDream](https://github.com/genekogan/deepdream), and [many others](https://github.com/ml4a/ml4a/tree/master/ml4a/models/submodules).

* a collection of [Jupyter notebooks](https://github.com/ml4a/ml4a-guides/tree/ml4a.net/examples) explaining the basics of deep learning for beginners, and providing [recipes for using the materials creatively](https://github.com/ml4a/ml4a-guides/tree/ml4a.net/examples/models).

## Example

ml4a bundles the source code of various open source repositories as [git submodules](https://github.com/ml4a/ml4a-guides/tree/ml4a.net/ml4a/models/submodules) and contains wrappers to streamline and simplify them. For example, to generate sample images with StyleGAN2:

```

from ml4a import image

from ml4a.models import stylegan

network_pkl = stylegan.get_pretrained_model('ffhq')

stylegan.load_model(network_pkl)

samples = stylegan.random_sample(3, labels=None, truncation=1.0)

image.display(samples)

```

Every model in `ml4a.models`, including the `stylegan` module above, imports all of the original repository's code into its namespace, allowing low-level access.

## Support ml4a

### Become a sponsor

You can support ml4a by [donating through GitHub sponsors](https://github.com/sponsors/ml4a/). 

### How to contribute

Start by joining the [Slack](https://join.slack.com/t/ml-4a/shared_invite/enQtNjA4MjgzODk1MjA3LTlhYjQ5NWQ2OTNlODZiMDRjZTFmNDZiYjlmZWYwNGM0YjIxNjE3Yjc0NWVjMmVlZjNmZDhmYTkzZjk0ZTg1ZGM%3E) or following us on [Twitter](https://www.twitter.com/ml4a_). Contribute to the codebase, or help write tutorials.

## License

ml4a itself is [licensed MIT](https://github.com/ml4a/ml4a/blob/master/LICENSE), but you are also bound to the licenses of any [models](https://github.com/ml4a/ml4a/tree/master/ml4a/models/submodules) you use.