1. Multi-Task-Based Spatiotemporal Generative Inference Network: A Novel Framework for Predicting the Highway Traffic Speed (MT-STGIN) [codes link]
- Note that, for the MT-STGIN program, should be reproduced on the TensorFlow version >= 1.13.1. If not, there are some problems will appear in the running stage.
- For me, I encountered the problem that is "could not initialize a memory primitive descriptor" when the TensorFlow version == 1.12.0.
2. Speed prediction of spatio-temporal attention network for dynamic highway network (ST-ANet) [codes link]
- This program has been accepted by the Journal of Computer Engineering. Therefore, you can download the codes on your computer device and use them as your component of research or subject.
3. How to Accurately Predict Traffic Speed Using Simple Input Variables? A Novel Self-Supervised Spatio-Temporal Bilateral Learning Network (3S-TBLN) [codes link]
- Note that, 3S-TBLN can be reproduced on the TensorFlow version >= 1.13.1, even the version >2.0.0. Then, we will upload all baselines' codes on this page.
--->>>>>>> MT-STGIN has been accepted by Expert Systems with Applications (a short conference version has been accepted by IEEE ITSC 2023);
--->>>>>>> 3S-TBLN has been submitted to IEEE Transactions on Intelligent Transportation Systems TITS (a short conference version has been submitted to IEEE ITSC);
--->>>>>>> ST-ANet has been accepted in Computer Engineering.
The above codes have been uploaded to the Git-Hub page [codes link].