Skip to content

About Fastkit-Mobility

Fastkit-Mobility is a Python library for human-mobility analysis. It pairs a familiar dataframe-facing API with Rust kernels for the compute-heavy parts of trajectory processing and measurement. See Features for what it can do.

PyPI Python License GitHub

Released 2026-10-28 (release notes) · Alpha · pandas and Polars via Narwhals

Relationship to scikit-mobility

Fastkit-Mobility follows the measure API of scikit-mobility and reimplements it on a Rust core with Narwhals dataframe support. Comparison tests against scikit-mobility run wherever the APIs overlap, and it is up to 2658x faster on 4M Brightkite rows (see benchmarks). Beyond that, Fastkit-Mobility has its own trajectory hierarchy, models and network tools.

License and credits

Fastkit-Mobility is released under the BSD 3-Clause license. Third-party code and algorithms, including MoveTK (Apache-2.0), are credited in THIRD_PARTY_NOTICES.md.

Cite Fastkit-Mobility

If you use Fastkit-Mobility in research, please cite the paper (HAL hal-05755990):

@inproceedings{santos2026fastkit,
  author    = {Santos, Gustavo H. and
               Viana, Aline Carneiro and
               Pappalardo, Luca and
               Silva, Thiago H.},
  title     = {Fastkit-Mobility: A
               High-Performance,
               DataFrame-Agnostic Library
               for Mobility Analysis},
  booktitle = {NetMob 2026},
  address   = {Niter{\'o}i, Brazil},
  year      = {2026},
  month     = oct,
  url       = {https://inria.hal.science/
               hal-05755990}
}

To cite a specific release of the software, use:

@software{fastmob,
  author  = {Santos, Gustavo H.},
  title   = {Fastkit-Mobility},
  note    = {Fast human-mobility
             analysis in Python.
             PyPI package fastmob},
  year    = {2026},
  version = {0.3.4},
  url     = {https://github.com/
             gefgu/fastmob}
}

Maintainer

Gustavo Henrique Bruno dos Santos (@gefgu).

Contribute and report issues

A useful bug report includes the Fastkit-Mobility and Python versions, the dataframe backend and its version, a minimal input, the expected and actual output, and the full traceback.