Table 3 Performance comparison between the CNN models in terms of the composition of the loss function.

Model	Loss function	RMSE of local maxima and minima		R² score of local maxima and minima
Model	Loss function	Train	Test	Train	Test
CNN	RMSE	0.0407	0.0702	0.9589	0.8768
	RMSE + diff. RMSE	0.0370	0.0703	0.9661	0.8767
	RMSE + diff. BCE	0.0413	0.0694	0.9576	0.8797
MLP	RMSE	0.0709	0.0914	0.8753	0.7913
Random forest	RMSE	0.0431	0.0894	0.9539	0.8006
Extra tree	RMSE	0.0288	0.0954	0.9794	0.7727

Root mean squared errors and R² scores were measured from local maxima and local minima of datasets.

Quick links

Search