Predicting the surface age of chondritic S-type asteroids using the space weathering features in reflectance spectra: Small data machine learning

Open Access

Table 3

Percentile factor error and RMSE for the two models.

Surface age		Number of samples	Ensemble model			GP model
Years	Log₁₀ (years)	Number of samples	RMSE log₁₀ (Years)	90^th percentile	95^th percentile	RMSE log₁₀ (Years)	90^th percentile	95^th percentile
Overall			0.291	2.95	3.87	0.247	2.39	2.89
10⁰−10²	1-2	20	0.466	3.1	5.2	0.452	2.9	5.6
10²−10³	2-3	20	0.204	3.1	5.2	0.176	2.9	5.6
10⁷−10⁸	7-8	6	0.543			0.337
10⁸−10⁹	8-9	88	0.241	2.9	3.6	0.179	2.2	2.4
10⁹−10¹⁰	9-10	34	0.304			0.198

Notes. RMSE = Root-mean-square error, GP = Gaussian process.

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.