K-Means Clustering

The above shows K-Means Clustering of 1000 randomly sampled points with k=3. Each point & the initial centroid are sampled from a normal distribution of X~N(0, 100*100). The points are iteratively reassigned to the cluster of the nearest centroid; The coordinates of the centroids are then calculated using the average of all its cluster members.

Usage

1. Installing packages

   pip install -r requirements.txt

2. Launching visualization

   python visualize.py

Contact

Chau Yuan Qi - @chauyuanqi - yuanqichau@gmail.com