Table 2.
Means of metrics of ten individual Zoobot fine-tuning runs, assuming a complete binary class split (non-merger class (class 0): merger probability < 0.5, merger class (class 1): merger probability > 0.5).
Class | Precision | Recall | F1-Score |
---|---|---|---|
Non-merger | 0.74 | 0.83 | 0.75 |
Merger | 0.80 | 0.70 | 0.80 |
Notes. Each run split the TNG50 data-set into different 63% training, 27% validation, and 10% testing data-sets. The metrics are based on the validation data-set.
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.