Table 1
Number of sources used to build each classifier and to derive its performance metrics.
| Samples with no flag | Samples with flags | |||||
| Studies | Training (70%) and test (30%) | Targets | Test | Targets | ||
| Case A (HL) | blazars (1572) | pulsars (134) | UnIds (422) | blazars (145) | pulsars (32) | UnIds (109) |
| Case A (LL) | blazars (1572) | galactic sources (183) | UnIds (169) | blazars (145) | galactic sources (75) | UnIds (247) |
| Case B | BL Lacs (638) | FSRQs (448) | BCUs (486) | – | – | – |
Notes. The number of sources to which each classifier was applied is also given: these “targets” correspond to unassociated sources (UnIds) for case A or blazars of unknown type (BCUs) for case B. HL and LL refer to the high and low galactic latitude studies (case A), respectively. Only BCUs are counted in the target sample for case B, the corresponding classifiers will be also applied to blazar candidates from case A.
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.