Table 1.

Performance comparison between ANN models, with and without the DCLDL loss, on the IRIS dataset

Class	ANN		ANN $+$ DCLDL
	Train Acc $(\%)$	Test Acc $(\%)$	Train Acc $(\%)$	Test Acc $(\%)$
0 (Setosa)	100.0	100.0	100.0	100.0
1 (Versicolor)	95.0	80.0	97.5	90.0
2 (Virginica)	100.0	100.0	100.0	100.0