A Joint Optimization of Momentum Item and Levenberg-Marquardt Algorithm to Level Up the BPNN’s Generalization Ability

Table 3

The results of different training methods for wine data.

Real labels	Preprocessed output			Postprocessed output
	ELM-BPNN	NLM-BPNN	Proposed LM-BPNN	ELM-BPNN	NLM-BPNN	Proposed LM-BPNN
1	1.00009249	1.00020181	1.00008609	1	1	1
1	0.99999165	0.99970850	1.00011684	1	1	1
1	1.00004976	1.00013716	1.00007315	1	1	1
1	1.00009044	1.06945969	1.05066960	1	1	1
1	0.99992444	1.01434834	1.01257733	1	1	1
1	0.99999324	1.00008633	1.00014440	1	1	1
2	2.00031311	2.00049736	2.00020179	2	2	2
2	1.80813014	2.02180286	2.03638972	2	2	2
2	2.00003265	1.99999689	1.99998956	2	2	2
2	1.99997582	2.00001218	2.00002534	2	2	2
2	0.40518282	1.99552685	1.98939772	0	2	2
2	1.99999567	1.99998810	1.99994678	2	2	2
2	1.99999881	2.00000198	2.00001719	2	2	2
2	1.99992264	2.00011454	1.99995018	2	2	2
3	2.98399414	2.68343848	2.52157921	3	3	3
3	2.99991935	3.00078223	2.99958012	3	3	3
3	3.00001821	2.99949485	3.00003921	3	3	3
3	3.00001865	2.99820215	2.99558228	3	3	3
3	2.56607392	2.32976636	2.77938018	3	2	3