v0.2.0 - Major Pipeline Optimization and Infrastructure Improvements

Major Changes

Performance Optimizations

• Migrate from NetworkX to rustworkx for 5-10x faster graph operations
• Replace shapefiles/GeoJSON with GeoParquet for 2-5x faster I/O operations
• Replace CSV with Parquet for tabular data storage
• Implement single Dijkstra algorithm (replaces multiple ego_graph calls)
• Add bounded Dijkstra algorithm for efficient walk time computation with distance bounding

Infrastructure Improvements

• DVC to Git LFS migration and automated data update system
• Automated data source updates with version checking and metadata tracking
• Data validation scripts for schema and data quality checks
• Graph conversion caching for improved performance

Testing & Development

• Comprehensive testing framework with pytest and pytest-cov
• Enhanced project dependencies including statsmodels for statistical analysis
• Parallel processing support for walk time calculations
• Migration scripts for converting existing data files to new formats

Documentation

• Data dictionary (DATA_DICTIONARY.md) with comprehensive details on data files and workflows
• Project backlog (BACKLOG.md) for tracking technical debt and feature requests
• CEJST workflow documentation (README_CEJST.md)
• H3 implementation details and Census API key setup instructions
• Enhanced README with updated guidance

Pipeline Enhancements

• Enhanced pipeline scripts (run_pipeline.sh, run_pipeline.py) with logging and error handling
• Jupyter notebooks for walk times and merging analysis
• Updated validation to support Parquet files
• Improved code organization and consistency across modules