Skip to main content
. 2022 Mar 3;13:706042. doi: 10.3389/fpls.2022.706042

TABLE 3.

The performance statistics of support vector regressor (SVR), extreme gradient boosting (XGB), deep neural network (DNN), and random forest (RF) models for lettuce.

Model Scenario SI T U95 MBE
SVR 1 0.035 0.647 31.90 1.59
2 0.032 0.015 29.35 0.034
3 0.029 1.600 26.10 3.10
XGB 1 0.051 0.780 46.80 2.84
2 0.031 0.110 28.70 –0.25
3 0.027 0.540 24.80 1.04
DNN 1 0.037 0.630 34.50 1.70
2 0.033 1.650 30.30 3.80
3 0.035 1.630 31.90 3.95
RF 1 0.039 0.160 36.2 –0.45
2 0.035 0.135 32.1 –0.34
3 0.033 0.087 30.3 –0.21

SI, Scatter Index; Tstat, T-statistic test; U95, Uncertainty with a 95% confidence level; MBE, mean bias error.