Table 1 Median statistics over the full 10-fold cross validation.

From: Predicting the inhibition efficiencies of magnesium dissolution modulators using sparse machine learning models

No. of features

3

5

63

1260

Model type

Tiny model

Small model

Medium model

Large model

Selection method

a

b

c

a

b

c

a

b

c

 

RMSE/pp

66

56

66

57

50

66

50

50

56

63

R2

0.40

0.51

0.30

0.68

0.66

0.41

0.60

0.62

0.50

0.35

Pearson’s r

0.61

0.71

0.54

0.82

0.81

0.64

0.77

0.79

0.70

0.59

p-value

0.23

0.13

0.27

0.05

0.06

0.17

0.08

0.06

0.12

0.22

  1. Median values of root mean squared errors (RMSE), coefficients of determination (R2), correlation coefficients (Pearson’s r) and p-values of the full 10-fold cross validation for all trained models by model type and feature selection method (a: ANOVA, b: RFE, c: random selection).
Back to article page