Skip to content

Resources

Datasets, packages, community and background reading for Fastkit-Mobility.

PyPI →

pip install fastmob

Source code →

Rust kernels, Python API and tests.

Get help →

Report bugs and ask questions.

Datasets

Brightkite check-ins scattered over the world map next to GeoLife GPS traces forming the ring roads of Beijing.
  • Brightkite: about 4.7M location check-ins from a social network (the benchmarks use the first 4M). The workload for the benchmarks.
  • GeoLife: GPS trajectories of 182 users in Beijing. Read it with load_geolife_trajectories; used in the stay-point notebook.
  • Foursquare NYC and others: datasets from scikit-mobility, available through load_dataset. See the data reference.

Check each dataset's own page for its license and terms of use before redistributing it.

Background reading

  • scikit-mobility: the library whose measure API Fastkit-Mobility follows.
  • trackintel: the staypoint, tripleg, trip and tour hierarchy that inspired Fastkit-Mobility's.
  • Narwhals: the dataframe-agnostic layer that lets Fastkit-Mobility accept pandas and Polars.

Papers

  • Santos, Viana, Pappalardo, Silva. Fastkit-Mobility: A High-Performance, DataFrame-Agnostic Library for Mobility Analysis. NetMob 2026 (the paper to cite for this library).
  • Pappalardo, Simini, Barlacchi, Pellungrini. scikit-mobility: A Python Library for the Analysis, Generation, and Risk Assessment of Mobility Data. Journal of Statistical Software, 2022.
  • Martin, Hong, Raubal and others. Trackintel: An open-source Python library for human mobility analysis. Computers, Environment and Urban Systems, 2023.
  • Cho, Myers, Leskovec. Friendship and mobility: user movement in location-based social networks. KDD 2011 (the Brightkite data).
  • Zheng, Zhang, Xie, Ma. Mining interesting locations and travel sequences from GPS trajectories. WWW 2009 (the GeoLife data).

Terms used in these docs

Ordered along the hierarchy from raw fixes to tours.

Positionfixes produce Staypoints and Triplegs, which combine into Trips and then Tours; Staypoints cluster into Locations, which tours return to.
Class Meaning
TrajDataFrame Trajectory: timestamped positions of one or many users
Positionfixes Position fix: one timestamped latitude/longitude observation
Staypoints Staypoint: a period spent within a small spatial area
Locations Location: a recurring place formed by clustering staypoints
Triplegs Tripleg: movement between two stays
Trips Trip: triplegs merged between two activity stays
Tours Tour: consecutive trips that return to the same place
FlowDataFrame OD flow: a count or value from an origin to a destination

Citing Fastkit-Mobility

See Cite Fastkit-Mobility for a BibTeX entry.