The Space team has made the following datasets publicly available.
Python and Weka command-line pipeline that builds the accessibility_score target, selects features, and compares regression and clustering models for the Project Methodology assignment.
Correlation matrix, PCA, clustering, and model-comparison figures including real Weka Explorer and Experimenter screenshots supporting the Project Methodology write-up.
Feature-selected regression and clustering tables, the accessibility_score target variable, and raw Weka output from the Project Methodology assignment.
Python pipeline used to fetch, clip, and aggregate all raw data sources into the final dataset.
Python, Weka Visualize and Preprocess screenshots supporting the outlier, correlation, balance, spatial-trend, and equity findings in the write-up.
The 46 DART station walksheds, the engineered feature table, the Census overlay, and the final combined flat file used for modeling.
GTFS transit feed, OpenStreetMap extracts (crosswalks, trees, benches, sidewalks, shops, resources), Dallas zoning polygons, and Census block group/tract data before spatial aggregation.
Powered by Clowder (1.22.1#1085 branch:master sha1:f28c203c56b2d4690d32ea0bce5364458de1ec79).