HKUDS/UrbanGPT
Spatio-temporal large language model for urban prediction tasks published at KDD 2024.
Not currently ranked — collecting fresh signals.
star history
UrbanGPT adapts LLMs to spatio-temporal prediction tasks in urban computing by combining temporal modeling with instruction tuning. The project provides PyTorch implementations, model weights on Hugging Face, and datasets for training spatio-temporal LLMs. It addresses urban challenges such as crowd flow prediction and traffic forecasting using foundation models.
Frequently asked
- What is HKUDS/UrbanGPT?
- Spatio-temporal large language model for urban prediction tasks published at KDD 2024.
- Is UrbanGPT open source?
- Yes — HKUDS/UrbanGPT is open source, released under the Apache-2.0 license.
- What language is UrbanGPT written in?
- HKUDS/UrbanGPT is primarily written in Python.
- How popular is UrbanGPT?
- HKUDS/UrbanGPT has 456 stars on GitHub.
- Where can I find UrbanGPT?
- HKUDS/UrbanGPT is on GitHub at https://github.com/HKUDS/UrbanGPT.