A machine learning approach to estimate mid-infrared fluxes from WISE data

Open Access

Table 2.

Goodness of fit metrics for the different ML models obtained during the ten-fold cross-validation step.

Metric	ML	MIPS1	MIPS1	IRAC4	IRAC4
	model	clean	unclean	clean	unclean
MAE ^(a)	ET	0.106	0.117	0.048	0.063
	LR	0.171	0.188	0.091	0.106
RMSE ^(a)	ET	0.165	0.179	0.083	0.110
	LR	0.238	0.254	0.139	0.160
R²	ET	0.940	0.945	0.983	0.973
	LR	0.873	0.890	0.952	0.944
MAPE ^(a)	ET	0.031	0.033	0.018	0.022
	LR	0.048	0.051	0.033	0.036

Notes.

^(a)

The values corresponding to these metrics refer to log transformed fluxes, not actual fluxes.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.