Chapter 10 builds a perceptron: a single neuron that learns a linear boundary. This expansion trains it on real measurements instead of synthetic points.
The page pulls a full year of daily weather for New York from Open-Meteo’s archive: mean humidity, mean pressure, and precipitation. Each day becomes a point placed by humidity and pressure and labeled by whether it rained. The perceptron runs one epoch per frame, adjusting three weights with the classic learning rule, and the dark line is the decision boundary the weights imply. You can watch the accuracy climb and settle, and see that humidity and pressure only get you so far, because rainy and dry days genuinely overlap. That overlap is the honest lesson: a single neuron can only draw a straight line, and some problems need more.
Data source: Open-Meteo Archive API
Endpoint:https://archive-api.open-meteo.com/v1/archive?latitude=40.71&longitude=-74.01&start_date=2025-01-01&end_date=2025-12-31&daily=precipitation_sum,relative_humidity_2m_mean,pressure_msl_mean
Access: Public, no key. Loaded from an hourly server-side refresh when possible, with a bundled snapshot fallback, and a live browser fetch only when the API allows CORS. Data window: Jan 1, 2025 - Dec 31, 2025 (pulled Jul 18, 2026 UTC)
Points drawn gray are the ones the current boundary gets wrong.