auto-clustering 0.1.0.dev0

Automatic Clustering selection with Ray Tune

auto-clustering

Automatic Clustering selection with Ray Tune

Important Note:

This tool may optimize metrics such as the validity index, silhouette or the davies-bouldin score, but as this is based on unsupervised learning, these metrics may not always reflect the true usefulness of the resulting clusters.

Example: Clustering Selection

from autoclustering import AutoClustering
from sklearn.datasets import load_digits


data, _ = load_digits(return_X_y=True)

clustering = AutoClustering(num_samples=50,
                            metric='validity_index',
                            n_jobs=-1,
                            verbose=0)

clustering.fit(data)

clustering.best_params_
clustering.best_score_
clustering.n_clusters_
clustering.best_estimator_

clustering.predict(data)

Changelog

See the changelog for notes on the changes of auto-clustering

Important links

• Official source code repo: https://github.com/rodrigo-arenas/auto-clustering/

• Download releases: https://pypi.org/project/auto-clustering/

• Issue tracker: https://github.com/rodrigo-arenas/auto-clustering/issues

• Stable documentation: https://auto-clustering.readthedocs.io/en/stable/

Source code

You can check the latest development version with the command:

git clone https://github.com/rodrigo-arenas/auto-clustering.git

Install the development dependencies:

pip install -r requirements.txt

Check the latest in-development documentation: https://auto-clustering.readthedocs.io/en/latest/

Contributing

Contributions are more than welcome!

There are several opportunities on the ongoing project, so please get in touch if you would like to help out. Make sure to check the current issues and also the Contribution guide.

Big thanks to the people who are helping with this project!

Testing

After installation, you can launch the test suite from outside the source directory:

pytest autoclustering/

Disclaimer

The library is still experimental and under heavy development