TABLE 1.

Parameters for PLS-DA models in parts and regions discrimination based on three levels of data sets.

Data	Model	LVs	R2	Q2	RMSEE	RMSECV	RMSEP	Accuracy (%)
								Training set	Test set
Parts	PLS-DA-L	1	0.198	0.159	0.374135	0.37687	0.295164	51.52	55.56
	PLS-DA-M	11	0.899	0.831	0.143237	0.167712	0.0758287	99.39	100
	PLS-DA-H	11	0.918	0.887	0.120499	0.138129	0.0669199	99.39	100
Regions	PLS-DA-L	/	/	/	/	/	/	/	/
	PLS-DA-M	14	0.584	0.333	0.349237	0.441024	0.325103	87.92	88.46
	PLS-DA-H	20	0.698	0.544	0.266242	0.351347	0.266231	95.34	92.22