Skip to main content

Table 2 Prediction performance using different input sequence lengths

From: Improving air pollutant prediction in Henan Province, China, by enhancing the concentration prediction accuracy using autocorrelation errors and an Informer deep learning model

Encoder Length

MAE

RMSE

Encoder Length

MAE

RMSE

6

5.66

9.63

28

5.75

9.69

8

5.57

9.47

30

5.73

9.70

10

5.67

9.63

32

5.76

9.73

12

5.65

9.67

34

5.76

9.74

14

5.67

9.62

36

5.73

9.65

16

5.66

9.59

38

5.75

9.67

18

5.63

9.58

40

5.74

9.63

20

5.61

9.54

42

5.70

9.63

22

5.68

9.66

44

5.71

9.63

24

5.71

9.70

46

5.70

9.66

26

5.70

9.64

48

5.75

9.69