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

Table 5

The basic information of test dataset and performances of the three networks.

	Number of tuples	ELM-BPNN		NLM-BPNN		Proposed LM-BPNN
		P1	P2	P1	P2	P1	P2
Apr.	83853	0.001648	2.457953𝑒−007	0.001637	2.423426𝑒−007	0.001637	2.423272𝑒−007
May	87518	0.001650	2.489964𝑒−007	0.001363	1.681887𝑒−007	0.001356	1.664438𝑒−007
Jun.	84380	0.001594	2.321456𝑒−007	0.001573	2.260457𝑒−007	0.001573	2.260597𝑒−007
Jul.	87222	0.001491	2.002518𝑒−007	0.001418	1.806153𝑒−007	0.001413	1.792959𝑒−007
Aug.	82798	0.001621	2.415615𝑒−007	0.001303	1.544337𝑒−007	0.001295	1.526117𝑒−007
Sep.	82473	0.001649	2.469237𝑒−007	0.001347	1.631628𝑒−007	0.001336	1.603953𝑒−007
Oct.	85797	0.001534	2.153020𝑒−007	0.001504	2.067444𝑒−007	0.001504	2.067388𝑒−007
Nov.	85131	0.001469	1.968304𝑒−007	0.001440	1.890057𝑒−007	0.001436	1.878854𝑒−007
Dec.	81871	0.001682	2.544444𝑒−007	0.001682	2.543521𝑒−007	0.001682	2.543497𝑒−007