Traffic Prediction on PeMSD4

13개 결과 · ⬇ CSV · JSON

12 steps MAE 낮을수록 좋음

Rank	Model	12 steps MAE	12 steps MAPE	12 steps RMSE	Extra Training Data	Paper	Code	Year
1	Hierarchical-Attention-LSTM (HierAttnLSTM)	9.168	–	22.844		Network Level Spatial Temporal Traffic State Forecasting with Hierarchical Attention LSTM (HierAttnLSTM)	TeRyZh/Network-Level-Travel-Prediction-Hierarchical-Attention-LSTM	2022
2	STD-MAE	17.80	–	–	✓	Spatial-Temporal-Decoupled Masked Pre-training for Spatiotemporal Forecasting	jimmy-7664/std-mae	2023
3	HTVGNN	17.99	–	–		A novel hybrid time-varying graph neural network for traffic flow forecasting		2024
4	PDG2Seq	18.24	12.09	30.08		PDG2Seq: Periodic Dynamic Graph to Sequence Model for Traffic Flow Prediction	wengwenchao123/PDG2Seq	2024
5	PDFormer	18.32	–	–		PDFormer: Propagation Delay-Aware Dynamic Long-Range Transformer for Traffic Flow Prediction	BUAABIGSCity/PDFormer	2023
6	PM-DMNet(P)	18.34	12.05	30.36		Pattern-Matching Dynamic Memory Network for Dual-Mode Traffic Prediction	wengwenchao123/PM-DMNet	2024
7	PM-DMNet(R)	18.37	12.01	30.68		Pattern-Matching Dynamic Memory Network for Dual-Mode Traffic Prediction	wengwenchao123/PM-DMNet	2024
8	DDGCRN	18.45	–	–		A Decomposition Dynamic graph convolutional recurrent network for traffic forecasting	wengwenchao123/DDGCRN	2023
9	FasterSTS	18.49	–	–		FasterSTS: A Faster Spatio-Temporal Synchronous Graph Convolutional Networks for Traffic flow Forecasting	Salabh2k2/Awesome-Project-Collection	2025
10	STWave	18.50	–	–		When Spatio-Temporal Meet Wavelets: Disentangled Traffic Forecasting via Efficient Spectral Graph Attention Networks	lmissher/stwave	2023
11	HAGCN	18.70	–	–		HAGCN : Network Decentralization Attention Based Heterogeneity-Aware Spatiotemporal Graph Convolution Network for Traffic Signal Forecasting		2022
12	STG-NRDE	19.13	12.68	30.94		Graph Neural Rough Differential Equations for Traffic Forecasting	jeongwhanchoi/STG-NCDE · jeongwhanchoi/stg-nrde	2023
13	STG-NCDE	19.21	12.76	31.09		Graph Neural Controlled Differential Equations for Traffic Forecasting	jeongwhanchoi/STG-NCDE	2021