Open Access

Table 4

Peak identification quality metrics for approximation models applied to the PLAsTiCC.

Model RMSE, days MAE, days RSE RAE MAPE, %
GP 3.14 ± 0.09 2.04 ± 0.04 0.0102 ± 0.0003 0.0075 ± 0.0002 0.0034 ± 0.0001
MLP (sklearn) 4.2 ± 0.1 2.67 ± 0.06 0.0136 ± 0.0003 0.0098 ± 0.0003 0.0044 ± 0.0001
MLP (pytorch) 5.4 ± 0.1 3.61 ± 0.07 0.0174 ± 0.0004 0.0133 ± 0.0003 0.0060 ± 0.0001
BNN 6.1 ± 0.1 4.39 ± 0.08 0.0199 ± 0.0004 0.0162 ± 0.0004 0.0073 ± 0.0001
NF 5.8 ± 0.1 3.95 ± 0.08 0.0188 ± 0.0004 0.0146 ± 0.0004 0.0066 ± 0.0001

Notes. Direct approach refers to when the peak is determined using the sum of all passbands.

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.