What problem does it solve?
This Skill consolidates usage guidance for the TransBigData Python library to reduce the friction of cleaning, processing, aggregating, and visualizing transportation spatiotemporal datasets such as GPS trajectories, taxi records, and transit network data.
Core Features & Use Cases
- Data preprocessing & quality checks: sampling interval analysis, boundary filtering, and ID reindexing to prepare raw GPS records.
- Trajectory processing: drift removal, redundancy cleaning, smoothing, segmentation, stay/move detection, densify/sparsify, and map-matching for robust trajectory analysis.
- Taxi OD extraction & trajectory separation: extract pickup/dropoff OD pairs and separate occupied vs idle trajectories for demand analysis.
- Gridding & aggregation: generate rectangular/hexagonal grids, geohash support, grid parameter optimization, and grid-to-area mapping for spatial aggregation.
- Coordinate conversion & distance tools: convert between WGS84, GCJ02, BD09 and compute distances reliably.
- Transit network modeling & visualization: build metro/bus networks, compute shortest paths, retrieve isochrones, and create heatmaps or OD flow maps for reporting.
Quick Start
Use the transbigdata skill to clean your GPS CSV, extract taxi OD pairs, convert coordinates to WGS84, aggregate results into 500m grids, and produce a heatmap.
