Fig. 10.
Download original image
Recall vs. the fraction of the training sample used per class. We notice a significant improvement for BeBRs and YSGs with increased training samples. When the samples sizes are already adequate, the maximum possible value is achieved faster (e.g., for RSGs and BSGs). The GAL and WR classes show an increase, while the LBV sample is too small to produce meaningful results.
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.