Machine Learning Specialization · Unsupervised Learning, Recommenders, Reinforcement Learning

K-Means Clustering and Image Compression

A clustering milestone that implements K-Means from scratch, then uses it to compress an RGB image by replacing thousands of colors with centroid colors.

K-MeansCentroid assignmentCentroid updateUnsupervised clusteringImage compression

GitHub

What this project is about

K-Means is shown twice: first as clustering in a 2D feature space, then as color compression. The same centroid assignment/update loop becomes a visual compression algorithm when pixels are treated as RGB points.

Implementation focus

This project made unsupervised learning concrete. Reducing an image to 16 color centroids is easier to understand than only plotting clusters, and it shows how a simple algorithm can solve a practical compression problem.

Key results and artifacts

2D dataset: 300 samples, 3 clusters
Bird image: 128×128×3 reshaped to 16,384 pixels
Compressed to K=16 colors

What I learned

This made unsupervised learning visible: the same centroid logic can group points and also reduce image color complexity.

Milestone deep dive

What this project proves in my learning path.

A clustering milestone that implements K-Means from scratch, then uses it to compress an RGB image by replacing thousands of colors with centroid colors. The milestone is included because it shows a complete learning unit: a problem, a data representation, an implementation path, measurable outputs, and a lesson that later connects to larger AI systems.

Input layerfeature matrices and geometric structure

Model layerunsupervised algorithm or probabilistic model

Evaluation layertests, metrics, plots, or qualitative output

Artifact layerclusters, projections, compressed images, or outliers

Problem framing

What I was trying to solve

This page is not just a lab description. It explains how I transformed the assignment idea into an engineering story: what the model receives, what it optimizes, what output it should produce, and how the result can be inspected.

Implementation

What I coded or assembled

The implementation focus was the mechanism behind the result: functions, model blocks, training logic, preprocessing, evaluation, and the small technical details that make the notebook work rather than only look correct.

Concepts

Core concepts

K-MeansCentroid assignmentCentroid updateUnsupervised clusteringImage compression

These tags show the technical ideas this milestone added to my toolkit.

Learning result

What changed after it

This made unsupervised learning visible: the same centroid logic can group points and also reduce image color complexity.

Important outputs and evidence

2D dataset: 300 samples, 3 clusters
Bird image: 128×128×3 reshaped to 16,384 pixels
Compressed to K=16 colors

How I would explain it in an interview

I would describe this milestone by starting from the data representation, then explain the model or algorithm, then show what was measured. The important part is not only the final number; it is the ability to reason about why the method works, what assumptions it makes, and what limitations should be improved next.