Open pknut opened 6 years ago
Seems that sometimes it is RMSE and sometimes it is log loss - depends on pipeline stage/subproblem:
See minerva/whales/validation.py
SCORE_FUNCTIONS = {'localization': rmse_multi,
'alignment': rmse_multi,
'classification': log_loss_whales,
'end_to_end': log_loss_whales
}
Results for which subproblem you've pasted above?
The validation score (2.0059) is not equal to validation loss (1.01667) or the validation accuracy (0.77713). Similarly, test score is hard to interpret. How are these two scores calculated?