Tensorflow implementation of capsule network (CapsNet) for traffic prediction.
Built with Python 3.5, TF-slim, numpy, pandas
- Obtain spatio-temporal image representation of traffic speed data, which are time series data with spatial relationship
- Compare capsule network (CapsNet) and convolutional neural network (CNN) in a short-term traffic speed forecasting problem
- Evaluated with real data measured by induction loop detectors on the road segments of Santander city in 2016
Please cite the following paper if you find this code helpful:
Youngjoo Kim, Peng Wang, Yifei Zhu, and Lyudmila Mihaylova, "A Capsule Network for Traffic Speed Prediction in Complex Road Networks", in Proceedings of Sensor Data Fusion: Trends, Solutions, and Applications (SDF), Bonn, Germany, Oct 2018. Available: https://www.researchgate.net/publication/327798102_A_Capsule_Network_for_Traffic_Speed_Prediction_in_Complex_Road_Networks