Home▸AI & Machine Learning▸Lag Features, Rolling Windows and Cyclical Hour Encoding: Forecast Error and Unit-Circle Map

Lag Features, Rolling Windows and Cyclical Hour Encoding: Forecast Error and Unit-Circle Map

A top-down companion view of the lag, rolling-window and cyclical-encoding feature bench: the load curve with its lag markers, rolling window and next-hour forecast, the 24 hour dial as a unit circle with sin/cos projections, and a ranked chart of test RMSE for each feature set. Same shared model and numerical checks as the 3D version.

AI & Machine Learning2DAdvanced60 FPS📱 Mobile-adapted🟠 Blender⇄ 3D version
Lag Features, Rolling Windows and Cyclical Hour Encoding: Forecast Error and Unit-Circle Map ↗ Open standalone

Turn a noisy hourly electricity-load series into the columns a model actually learns from: lag features, rolling-window mean and spread, and a sine/cosine clock. A ridge regression forecasts the next hour and is scored on nine unseen days against persistence and the noise floor, with built-in numerical checks. Desk, window posts, pegs and dial modelled in Blender.

⚙ Under the hood

A top-down companion view of the lag, rolling-window and cyclical-encoding feature bench: the load curve with its lag markers, rolling window and next-hour forecast, the 24 hour dial as a unit circle with sin/cos projections, and a ranked chart of test RMSE for each feature set. Same shared model and numerical checks as the 3D version.

Machine LearningTime SeriesFeature EngineeringCyclical EncodingRidge Regression

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)