Street Light Scheduling (ANN + EA + Fuzzy)

This project generates a synthetic street-light dataset, produces exploratory visualizations, trains an ANN to predict hourly activity, and optimizes a 24-hour lighting schedule using evolutionary algorithms (standard EA, fuzzy EA, and tuned fuzzy EA).

Files

• data_gen.py — dataset generation → street_light_dataset.csv
• plots.py — creates 12 plots (01_*.png … 12_*.png) + PCA summary prints
• ann_model.py — ANN pipeline (preprocessing, optimizer comparison, final training, evaluation plots)
• ea_optimizer.py — schedule optimization + convergence/analysis plots
• main.py — runs the full pipeline end-to-end

Setup

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt