This repository is the official implementation of TSIN("Coupling Makes Better: An Intertwined Neural Network for Taxi and Ridesourcing Demand Co-Prediction") with Pytorch.
- NYC-Taxi & FHV: https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page
- CHI-Taxi: https://data.cityofchicago.org/Transportation/Transportation-Network-Providers-Trips-2018/rpfj-eb3a
- CHI-TNP: https://data.cityofchicago.org/Transportation/Transportation-Network-Providers-Trips/m6dm-c72p
Pre-processing of trip data will be released later...
- Pytorch >= 1.8
- Numpy
# train the TSIN model
run train.py
# test the TSIN model
run test.py