Table 4.
Confusion matrix and metrics of the classification applied to the reference sample of 2SXPS.
| Truth → | AGN | Star | XRB | CV | Total cl. |
|---|---|---|---|---|---|
| →AGN | 19515 | 82 | 25 | 191 | 19813 |
| →Star | 44 | 4628 | 3 | 27 | 4702 |
| →XRB | 140 | 18 | 326 | 17 | 501 |
| →CV | 9 | 9 | 2 | 124 | 144 |
| Total | 19708 | 4737 | 356 | 359 | Average |
| recall (%) | 99.0 | 97.7 | 91.6 | 34.5 | 80.7 |
| precision (%) | 97.0 | 98.6 | 90.7 | 85.5 | 92.3 |
| f1 score | .980 | .981 | .911 | .492 | .841 |
Notes. The first four rows show the number counts of objects classified as AGN, stars, XRBs, and CVs, respectively. The last three rows give the recall rate, the true positive rate (precision), and the f1 score associated with each class, and those averaged over all classes are given in the last column.
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.