KNN Color Proximity Project

This Project has been made with one prompt (qwen3.6:35b-a3b , with ollama) in less than an hour using claude code

A K-Nearest Neighbors (KNN) classifier that predicts color names from RGB values.

How it works

1. Dataset: 200+ named colors with their RGB values
2. KNN Engine: From-scratch implementation using Euclidean distance in RGB space
3. Prediction: Given RGB values, finds the K most similar colors and predicts the name

Project Structure

projet_knn/
├── app.py              # Flask backend
├── data/
│   └── colors.csv      # Color dataset (name, r, g, b)
├── src/
│   ├── __init__.py
│   └── knn.py          # KNN implementation
├── static/
│   ├── app.js          # Frontend logic
│   └── style.css       # Styling
├── templates/
│   └── index.html      # Main page
└── requirements.txt    # Dependencies

Setup

pip install -r requirements.txt
python app.py

Then open http://localhost:5000 in your browser.

Features

• Interactive color picker with RGB sliders and number inputs
• KNN prediction with confidence scores and distance metrics
• RGB space visualization showing the input color's position
• Random color button for quick testing
• Adjustable K parameter (1-20)