Geosnap

Provides fast tools to standardize multi-period data into shared geographic representations, essential for spatiotemporal analysis as highlighted in the main features.

Combines unsupervised ML from scikit-learn and spatial optimization from PySAL for advanced neighborhood structure analysis, including model diagnostics and outlier detection.

Offers quick access to a large U.S. neighborhood indicator database from Census, EPA, and others via quilt and geoparquet, streamlining data acquisition for researchers.

Includes novel techniques for identifying change hotspots and simulating future neighborhood conditions, directly supporting research on gentrification and suburbanization trends.