Total and dark mass from observations of galaxy centers with machine learning

Open Access

Table 4.

Accuracy as a function of the number of features for the M_tot(r_1/2), considering measurement errors.

R²	Features	num

MELA_NETG
0.8957	M_⋆	1
0.9544	r_1/2, M_⋆	2
0.9584	r_1/2, M_⋆, σ	3
0.9604	r, r_1/2, M_⋆, σ	4
0.9606	g, r, r_1/2, M_⋆, σ	5

MELA_DW

0.4242	σ	1
0.7139	r_1/2, M_⋆	2
0.8203	r_1/2, M_⋆, σ	3
0.8306	g, r_1/2, M_⋆, σ	4
0.8326	g, r, r_1/2, M_⋆, σ	5

MELA_LTG

0.7803	M_⋆	1
0.9405	r_1/2, M_⋆	2
0.9464	r_1/2, M_⋆, σ	3
0.9487	r, r_1/2, M_⋆, σ	4
0.9490	g, r, r_1/2, M_⋆, σ	5

MELA_ALL

0.8238	M_⋆	1
0.9419	r_1/2, M_⋆	2
0.9560	r_1/2, M_⋆, σ	3
0.9581	r, r_1/2, M_⋆, σ	4
0.9585	g, r, r_1/2, M_⋆, σ	5

Notes. Each row displays the optimal feature combination for the corresponding feature count. The results without considering measurement errors can be found in Table B.2.

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.