Table 3 Comparing ODEFormer with traditional time series models.

From: Construction of a traffic flow prediction model based on neural ordinary differential equations and Spatiotemporal adaptive networks

Traditional time series models

PEMS03

PEMS04

PEMS07

PEMS08

MAE

RMSE

MAPE (%)

MAE

RMSE

MAPE (%)

MAE

RMSE

MAPE (%)

MAE

RMSE

MAPE (%)

ARIMA

35.31

47.59

33.78

33.73

48.80

24.18

38.17

59.27

19.46

31.09

44.32

22.73

SVR

21.97

35.29

21.51

28.70

44.56

19.20

32.49

50.22

14.26

23.25

36.16

14.64

ODEFormer

14.37

25.62

14.25

18.31

29.66

12.51

19.48

32.14

8.25

13.85

22.88

9.05

\(\Delta {E_{\text{1}}}\)

20.94

21.97

19.53

15.42

19.14

11.67

18.69

27.13

11.21

17.24

21.44

13.68

\(\Delta {E_2}\)

7.6

9.63

7.26

10.39

14.9

6.69

13.01

18.08

6.01

9.4

13.28

5.59

\(\overline {{\Delta {E_N}}}\)

MAE:14.08 RMSE:18.20 MAPE:10.21%

\(\overline {{\Delta {P_N}}}\)

MAE:45.09% RMSE:39.14% MAPE:0.47%