Research Article

Machine Learning for Estimating Leaf Dust Retention Based on Hyperspectral Measurements

Table 2

Training and testing results for three plants using the different regression algorithms.

PlantsAlgorithmsNumber of bandsTrainingTesting
MAE (g/m2)RMSE (g/m2)BiasNumber of samplesMAE (g/m2)RMSE (g/m2)BiasNumber of samples

GLCART10000.980.040.080.00990.780.170.260.0542
940.950.080.130.000.760.180.260.04
RF10000.980.070.100.000.880.140.190.05
940.980.060.080.000.920.110.150.03
SVM10000.970.080.09−0.010.890.140.190.05
940.920.130.160.000.890.160.190.07
LCCART10000.970.050.100.00970.500.330.47−0.0641
1670.950.070.120.000.500.290.46−0.03
RF10000.970.080.110.000.830.170.23−0.03
1670.980.070.100.000.870.140.20−0.02
SVM10000.970.090.10−0.010.800.170.23−0.02
1670.950.110.130.000.770.200.25−0.01
CFCART10000.930.080.130.00940.640.200.310.0041
730.980.040.070.000.700.170.290.03
RF10000.970.060.090.000.800.160.230.02
730.980.050.080.000.840.140.210.02
SVM10000.970.060.080.010.900.130.170.03
730.970.080.080.000.880.140.18−0.01