Geomancer: An Open-Source Framework for Geospatial Feature Engineering

TL;DR

Geomancer is an open-source framework that streamlines geospatial feature engineering for large-scale machine learning, enabling easy feature creation, sharing, and reproducibility from spatial datasets.

Contribution

It introduces a flexible, user-friendly system for geospatial feature engineering that integrates with data warehouses and supports sharing via JSON.

Findings

Useful in property value estimation and area valuation tasks.

Simplifies large-scale geospatial feature extraction.

Enhances reproducibility and sharing of features.

Abstract

This paper presents Geomancer, an open-source framework for geospatial feature engineering. It simplifies the acquisition of geospatial attributes for downstream, large-scale machine learning tasks. Geomancer leverages any geospatial dataset stored in a data warehouse, users need only to define the features (Spells) they want to create, and cast them on any spatial dataset. In addition, these features can be exported into a JSON file (SpellBook) for sharing and reproducibility. Geomancer has been useful to some of our production use-cases such as property value estimation, area valuation, and more. It is available on Github, and can be installed from PyPI.