Free Access

Table 1.

Applied regression models with RMSEtest errors using all the attributes of galaxies and without the VLG value.

Model All Without SGB VLG log d25 bri25 U B I K UI BK
attributes VLG SGL
Linear 0.38 0.52 0.48 37 0.22 0.64 0.01 0 0 0.01 0.14 0.02
Polynomial 0.37 0.49 0.87 32 0.27 1.16 0.78 0.06 0 0.05 1.67 0.05
k-NN 0.37 0.50 0 35 0.01 0 0 0.01 0.02 0 0 0
Gradient boosting 0.36 0.44 1.51 22 0.18 0 0.04 1.02 0.95 0 0 0
ANN 0.35 0.44 1.09 26 0.33 0 0.16 0.01 0.47 0.11 0.71 0.22

Notes.

The last ten columns show the importance of each attribute for a given regression model, which is given as an increase in error as a percentage when leaving out a given attribute.

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.