Open Access

Table 4.

Summary of the performance of the CNN at identifying lens candidates regardless of deflector type.

Grade Mag. bin Candidates Candidates Completeness Purity of the Prevalence in Ratio of CNN-purity
found expected CNN sample parent sample to base prevalence
A + B 17 to 18 4 55 7% 7.7 × 10−3 1.7 × 10−4 46.2
A + B 18 to 19 13 100 13% 3.5 × 10−3 1.0 × 10−4 34.8
A + B 19 to 20 21 380 6% 2.3 × 10−3 1.3 × 10−4 17.0
A + B 20 to 20.5 8 187 4% 2.0 × 10−3 6.7 × 10−5 29.9

C 17 to 18 9 120 8% 1.7 × 10−2 3.7 × 10−4 47.3
C 18 to 19 22 200 11% 5.9 × 10−3 2.0 × 10−4 29.4
C 19 to 20 12 190 6% 1.3 × 10−3 6.7 × 10−5 19.5
C 20 to 20.5 1 374 0% 2.5 × 10−4 1.3 × 10−4 1.9

Notes. Table follows the layout of Table 3.

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.