Classification results of RBFNN models based on 3000 and 1100 samples of each variety in the calibration set.

Samples of each variety in the calibration set		Calibration				Prediction
		1	2	3	Accuracy	1	2	3	Accuracy
3000	1	2931	12	57	97.70%	1180	36	84	90.77%
	2	11	2961	28	98.70%	33	1210	57	93.08%
	3	45	24	2931	97.70%	38	15	1247	95.92%
	Total				98.03%				93.26%
1100	1	1070	5	25	97.27%	1192	25	83	91.69%
	2	7	1084	9	98.55%	42	1209	49	93.00%
	3	22	6	1072	97.45%	51	21	1228	94.46%
	Total				97.76%				93.05%