Table 4 The Performance of Baseline Approaches.

		Beijing to Guangzhou		Beijing to Guangzhou		Changsha to Foshan		Changsha to Foshan		Wuhan to Xian		Wuhan to Xian
		2021		2022		2021		2022		2021		2022
		MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE
Non-transfer	T-GCN	0.62	0.71	0.72	0.81	0.45	0.63	0.55	0.94	0.51	0.62	0.52	0.84
	ASTGCN	0.68	0.78	0.63	0.74	0.58	0.92	0.50	0.92	0.53	0.82	0.41	0.91
	LSTM	0.75	0.87	0.92	1.01	0.49	0.89	0.50	0.72	0.51	0.76	0.52	0.74
	XGBoost	0.51	0.60	0.41	0.62	0.42	0.88	0.37	0.65	0.51	0.84	0.33	0.64
	CNN-LSTM	0.66	0.61	0.51	0.82	0.42	0.78	0.39	0.66	0.43	0.62	0.49	0.62
	Transformer-based Model	0.75	0.66	0.48	0.69	0.48	0.89	0.30	0.75	0.49	0.73	0.36	0.65
transfer	MetaST	0.54	0.73	0.43	0.66	0.43	0.73	0.48	0.87	0.44	0.62	0.58	0.72
	RegionTrans	0.50	0.72	0.55	0.73	0.55	0.81	0.44	0.87	0.56	0.72	0.47	0.89
	TL_DCRNN	0.48	0.58	0.38	0.60	0.48	0.85	0.36	0.66	0.47	0.72	0.25	0.73
	DANN	0.40	0.66	0.42	0.64	0.58	0.87	0.40	0.86	0.54	0.89	0.47	0.76
	TransCSM(ours)	0.27	0.56	0.30	0.46	0.36	0.61	0.28	0.64	0.33	0.59	0.23	0.60
	TransCSM(w/o ConvGRU)	0.43	0.64	0.34	0.57	0.41	0.67	0.32	0.53	0.43	0.69	0.41	0.62
	TransCSM(w/o DA)	0.45	0.87	0.39	0.63	0.45	0.74	0.38	0.55	0.47	0.77	0.43	0.61

Quick links

Search