Table 7.

Results of fine-tuned models using different CNN architectures.

Source/target domain	Model	Calibration seta		Validation setb		Prediction setc
		R2	RMSE	R2	RMSE	R2	RMSE
LMY24/XLZ53	CNN1	0.921	2.336	0.853	3.217	0.880	2.903
	CNN2	0.909	2.505	0.850	3.248	0.870	3.020
	CNN3	0.910	2.501	0.821	3.550	0.855	3.184
	CNN4	0.892	2.739	0.840	3.356	0.855	3.186
	AlexNet	0.877	2.923	0.848	3.269	0.852	3.222
	VGGNet-9	0.897	2.672	0.819	3.570	0.840	3.348
XLZ53/LMY24	CNN1	0.891	2.691	0.828	3.399	0.820	3.476
	CNN2	0.907	2.454	0.828	3.397	0.818	3.494
	CNN3	0.910	2.444	0.836	3.319	0.818	3.497
	CNN4	0.898	2.599	0.826	3.414	0.817	3.508
	AlexNet	0.864	3.003	0.813	3.549	0.819	3.489
	VGGNet-9	0.891	2.693	0.813	3.550	0.816	3.509

^aCalibration set means the calibration set of the target domain; ^bvalidtion set means the validation set of the target domain; ^cprediction set means the prediction set of the target domain.