Table 1
Summary of the overall classification performance evaluated on the validation data set.
| Prior | Class | Overall completeness | Overall purity | ||||
|---|---|---|---|---|---|---|---|
| Original | Cleaned | Combmod-α | Original | Cleaned | Combmod-α | ||
| Quasar | 0.916 | 0.916 | 0.819 | 0.240 | 0.489 | 0.749 | |
| Galaxy | 0.935 | 0.953 | 0.931 | 0.219 | 0.353 | 0.590 | |
| Global prior | Star | 0.996 | 0.998 | 0.999 | 0.990 | 0.990 | 0.990 |
| White dwarf | 0.432 | 0.430 | 0.395 | 0.250 | 0.406 | 0.429 | |
| Binary | 0.002 | 0.002 | 0.0014 | 0.075 | 0.062 | 0.079 | |
Notes. Predicted labels are obtained from the class with the maximum posterior probability. The labels ‘original’ and ‘cleaned’ refer to the results evaluated on the full validation data set and the cleaned validation data set, respectively. Combmod-α is the parametric combination of the GDR3 DSC Allosmod and Specmod, where αglob = [1.0, 0.6, 0.5]. Results of the extragalactic classes are highlighted in boldface.
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.