Home

Machine learning utilities for GIS operations, including automated building footprint extraction and object classification from orthophoto imagery.

• Target Framework: net10.0

📖 Guides

• Year-Built-Prediction — what the model predicts and what it actually learns, the requirements for running it, and how to drive DiGi.GIS.YOLO.UI.ConsoleApp.exe.
• Training-the-Year-Built-Model — retraining step by step, from populating building_data through to recording provenance. Note that Model Builder cannot train this on Visual Studio 2026; the mlnet CLI can.

The year-built label is the first year a building appears in the orthophoto record, not a historical construction year — 84 % of labelled rows carry 2008. A trivial "first detection year" rule already scores R² 0.783, so read the baselines before trusting any headline metric.

🔗 Dependencies

• DiGi.Core

---

🌐 DiGi Ecosystem

• Foundational: DiGi.Core | DiGi.Math | DiGi.Unit | DiGi.Log
• Geometry & Graphics: DiGi.Geometry | DiGi.GLTF | DiGi.Rhino
• GIS & Data: DiGi.GIS | DiGi.OSM | DiGi.BDOT10k
• Simulation: DiGi.Analytical | DiGi.Solar | DiGi.Tas

Part of the DiGi software suite for BIM and CAD integrations.