Application of machine learning techniques to simulate the evaporative fraction and its relationship with environmental variables in corn crops

Ecological Processes

Table 4 Results of the evaluation criteria applied on the 4 ML algorithms selected during the training-and-testing phase

		R²	RMSE	MAE	MSE
Training phase
Machine learning model
Support vector machine	Polynomial SVM	0.76	0.0530	0.0388	0.001
	Cubic Gaussian SVM	0.83	0.0540	0.0472	0.002
Gaussian process regression	Matern 5/2	0.94	0.0293	0.0236	0.000
Gaussian process regression	Rational quadratic	0.99	0.0134	0.0106	0.000
Testing phase
Machine learning model
Support vector machine	Cubic Gaussian SVM	0.66	0.0806	0.0678	0.004
	Polynomial SVM	0.70	0.0741	0.0662	0.004
Gaussian process regression	Matern 5/2	0.82	0.0593	0.0458	0.002
Gaussian process regression	Rational quadratic	0.72	0.0554	0.0418	0.001