K-Means Clustering From Scratch | Dataford Interview Questions

Implement a function to perform k-means clustering from scratch.

Use squared Euclidean distance, assign ties to the lowest-indexed centroid, and stop when assignments stabilize or max_iters is reached. Empty clusters must retain their previous centroid.

Signature: def k_means(points, k, max_iters, initial_centroids):. Return (assignments, centroids), where assignments is a list of cluster indices and centroids is a list of coordinate lists.

Constraints

1 <= len(points) <= 1000
1 <= k <= len(points)
1 <= len(initial_centroids) = k
1 <= len(points[i]) <= 20
All points and centroids have the same dimensionality
1 <= max_iters <= 100
Coordinates are finite integers or floating-point values

Function Signature

def k_means(points, k, max_iters, initial_centroids):

Your question is K-Means Clustering From Scratch. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem