Table 2.
Sources used in this work.
| Survey | w/ | w/ | Unique |
|---|---|---|---|
| used for training | LS10 | spec. redshift | |
| 4XMM | 32 900 | 9881 | 4031 (27.7%) |
| eFEDS | 27 367 | 10 309 | 10 142 (69.6%) |
| HIGH-Z AGNs | 391 | 391 | 391 (2.7%) |
| Blind samples | |||
| CSC2 | 5643 | 1654 | 416 |
| Chandra-COSMOS | 2491 | 2198 | 1699 |
Notes. The table shows the breakdown distribution of sources used for training (namely: training, test, and validation samples) and in the blind samples, before and after the match with the spectroscopic redshift compilation, and after cleaning for duplicates. The numbers within the parenthesis show the contribution to the sample used for training.
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.