scikit-mobility/DeepGravity
a PyTorch implementation of the paper "Deep Gravity: enhancing mobility flows generation with deep neural networks and geographic information"
This project helps urban planners, public health officials, and transportation analysts understand how people move within and between cities, even when direct mobility data is unavailable. It takes geographic information about various facilities (like roads, businesses, hospitals) and location populations to predict the likelihood of travel between different points. The output is a realistic estimate of mobility flows, allowing users to make informed decisions about infrastructure, resource allocation, and public health strategies.
115 stars. No commits in the last 6 months.
Use this if you need to understand or predict human mobility patterns in a region where real-time or historical movement data is scarce or nonexistent.
Not ideal if you have extensive, high-fidelity mobility data available and require highly precise individual-level movement tracking rather than aggregate flow probabilities.
Stars
115
Forks
40
Language
Jupyter Notebook
License
—
Category
Last pushed
Dec 29, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/scikit-mobility/DeepGravity"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lucidrains/rectified-flow-pytorch
Implementation of rectified flow and some of its followup research / improvements in Pytorch
probabilists/zuko
Normalizing flows in PyTorch
davidnabergoj/torchflows
Modern normalizing flows in Python. Simple to use and easily extensible.
keishihara/flow-matching
Flow Matching implemented in PyTorch
LukasRinder/normalizing-flows
Implementation of normalizing flows in TensorFlow 2 including a small tutorial.