https://github.com/akapich/clustermatic

Python AutoML library for clustering tasks
https://github.com/akapich/clustermatic

automl clustering machine-learning scikit-learn

Last synced: 4 months ago
JSON representation

Python AutoML library for clustering tasks

• Host: GitHub
• URL: https://github.com/akapich/clustermatic
• Owner: AKapich
• Created: 2024-12-21T22:42:47.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-12-27T22:17:58.000Z (over 1 year ago)
• Last Synced: 2026-01-05T16:23:32.265Z (6 months ago)
• Topics: automl, clustering, machine-learning, scikit-learn
• Language: Python
• Homepage: https://pypi.org/project/clustermatic/
• Size: 826 KB
• Stars: 0
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
![clustermatic](https://raw.githubusercontent.com/AKapich/clustermatic/refs/heads/main/clustermatic/auxiliary/clustermatic.png)

---

![PyPI](https://img.shields.io/pypi/v/clustermatic)

![Python Versions](https://img.shields.io/pypi/pyversions/clustermatic)

`clustermatic` is a Python library designed to accelerate clustering tasks using `scikit-learn`. It serves as a quick tool for selecting the optimal clustering algorithm and its hyperparameters, providing visualizations and metrics for comparison.

## Features

- **Clustering Algorithms**: Analyzes six clustering algorithms from `scikit-learn`:

    - `KMeans`

    - `DBSCAN`

    - `MiniBatchKMeans`

    - `AgglomerativeClustering`

    - `OPTICS`

    - `SpectralClustering`

- **Optimization Methods**: Includes Bayesian optimization and random search for hyperparameter tuning.

- **Flexible Preprocessing**: Allows users to customize how the data is meant to be preprocessed, adjusting methods such as scaling, normalization, and dimensionality reduction.

- **Evaluation Metrics**: Supports evaluation with `silhouette`, `calinski_harabasz`, and `davies_bouldin` scores.

- **Report Generation**: Generates reports in HTML format after optimization.

## Installation

To install `clustermatic`, use pip:

```bash

pip install clustermatic

```

## Usage

For a quick start, use the following code snippet:

```python

from clustermatic import AutoClusterizer

# Load data

from sklearn.datasets import make_moons

X, _ = make_moons(n_samples=200, noise=0.1, random_state=42)

# Initialize AutoClusterizer

ac = AutoClusterizer()

# Fit the data

ac.fit(X)

# Generate report

ac.evaluate()

```

For more detailed walkthrough, check out [this example Jupyter Notebook](https://github.com/AKapich/clustermatic/blob/main/examples/example.ipynb)