This project is an implementation of the K-means clustering algorithm entirely hand-coded in Python, without using machine-learning libraries such as scikit-learn. The aim is to understand the fundamental concepts of the K-means algorithm and to provide a simple tool for learning and experimentation.
The K-means algorithm is a partitioning method that divides a data set into K distinct clusters, minimizing the intra-cluster variance. This algorithm is widely used in machine learning for unsupervised clustering tasks.
Markup :
- Customized initialization: Use a random or defined centroid initialization method.
- Number of clusters: Specify the number of clusters to be identified in your data.
- Multiple iterations: The model executes any number of iterations.
- Results display: Clustering results can be displayed graphically (2D or 3D).
Thanks to all those who contributed to the improvement of this project. This project was inspired by the desire to understand clustering algorithms in depth and to share this understanding with the community.