# LibCity/Bigscity-LibCity

LibCity: An Open Library for Urban Spatial-temporal Data Mining

Repository: https://github.com/LibCity/Bigscity-LibCity
Canonical: https://ross.abutalabs.com/products/bigscity-libcity
Homepage: https://libcity.ai/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: traffic, traffic-prediction, traffic-flow-prediction, traffic-speed-prediction, on-demand-service, trajectory-prediction, toolkit, deep-learning, spatio-temporal-prediction, spatio-temporal, pytorch-implementation, traffic-forecasting, map-matching, representation-learning, estimated-time-of-arrival, traffic-accident-prediction, od-matrix, time-series-prediction, eta, libcity
Last push: 2024-12-18T08:11:29+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2090, "days_push": 623, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1157, forks 193 (observed 2026-08-28T04:03:48.287239+00:00)

## What it is
LibCity is an open-source PyTorch library for urban spatial-temporal data mining, providing a unified pipeline for traffic prediction research. It reproduces 74 models across 9 tasks (traffic flow/speed prediction, ETA, map matching, etc.) with 52 datasets and standardized evaluation.

## Use cases
- predict traffic flow on road networks
- forecast traffic speed from sensor data
- estimate travel time between origin and destination
- predict next location from trajectory data
- map matching GPS points to road networks
- learn road network representations
- benchmark spatio-temporal prediction models
- predict on-demand service demand

## When to choose
- you need a unified framework to reproduce and compare traffic prediction models
- you want standardized datasets and evaluation metrics for spatio-temporal research
- you are a researcher prototyping new traffic prediction models in PyTorch

## When to avoid
- you need production traffic forecasting at scale rather than research experiments
- your task is general time-series forecasting unrelated to urban/spatial data
- you prefer TensorFlow or non-PyTorch ecosystems

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, data-science, benchmarking, data-visualization
- domain: machine-learning, deep-learning, data-science
- platform: python, cross-platform
- tags: traffic-prediction, spatio-temporal, trajectory-prediction, eta, map-matching, pytorch, urban-computing, research-toolkit, algorithms, gpu

## Member repositories
- LibCity/Bigscity-LibCity (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:48.287239+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:32:00.618202+00:00, confidence not recorded.
  - readme: https://github.com/LibCity/Bigscity-LibCity (fetched 2026-08-28T04:03:48.287239+00:00, sha fc04edc999e6)
  - homepage: https://libcity.ai/ (fetched 2026-08-29T12:36:53.576901+00:00, sha a4575ddca121)
- Data as of 2026-08-30T08:39:29.467469+00:00.