Cough event classification by pretrained deep neural network

Table 4 Result table with randomly initialized neural network on PD.

# layers	# hidden units	Sens.	Spec.	F1	Macro Ave.	Mic. Ave.
1	512	0.768	0.909	0.663	0.838	0.889
1	1024	0.769	0.908	0.661	0.839	0.888
1	1536	0.783	0.903	0.661	0.843	0.886
1	2048	0.784	0.902	0.66	0.843	0.885
2	512	0.746	0.914	0.659	0.83	0.89
2	1024	0.759	0.913	0.664	0.836	0.891
2	1536	0.744	0.912	0.653	0.828	0.888
2	2048	0.745	0.912	0.654	0.828	0.888
3	512	0.755	0.905	0.648	0.83	0.884
3	1024	0.745	0.912	0.654	0.829	0.888
3	1536	0.735	0.908	0.641	0.821	0.883
3	2048	0.746	0.909	0.65	0.828	0.886
4	512	0.759	0.897	0.637	0.828	0.877
4	1024	0.736	0.905	0.637	0.82	0.881
4	1536	0.744	0.901	0.636	0.823	0.879
4	2048	0.693	0.88	0.573	0.787	0.853

ISSN: 1472-6947