Identification of Dry Bean Varieties Based on Multiple Attributes Using CatBoost Machine Learning Algorithm

Table 7

10-Fold cross validation accuracy.

Fold nos	No. of instances (90 : 10)		Random forest	XGBoost	CatBoost
	Training sets	Test sets
1	3288	366	92.89	94.53	94.53
2	3288	366	91.25	93.16	95.35
3	3288	366	92.34	92.07	94.81
4	3288	366	93.44	91.25	92.89
5	3289	365	92.05	92.87	93.97
6	3289	365	93.42	93.15	93.42
7	3289	365	95.06	93.15	93.69
8	3289	365	92.32	90.95	92.87
9	3289	365	93.97	95.06	94.52
10	3289	365	91.78	92.05	92.05
Mean accuracy in percentage			92.9	92.8	93.8