WaywardHouse
Series Overview
This series is in development. Content will expand to cover the full range of supervised and unsupervised learning methods applied to geographic and environmental datasets.
Machine learning methods have become central tools in computational geography — not as replacements for process-based models, but as complements: pattern recognition at scales and resolutions that explicit models struggle to reach. This series develops those methods from mathematical first principles, with geographic applications throughout.
Current Models
Models in this series assume familiarity with Series 1 (Foundations) and Series 3 (Spatial Analysis).
Prerequisites
- Series 1: Foundations of Computational Geography
- Series 3: Computational Spatial Analysis
- Linear algebra (vectors, matrices, eigendecomposition) — introduced within the models as needed