Update to Lensing Challenge

ML4SCI / DeepLearnHackathon

3 stars 6 forks source link

Closed pranath-reddy closed 2 days ago

pranath-reddy commented 2 days ago

scipy.interpolate.interp1d is causing a divide-by-zero error, leading to NaN values in the calculation of the macro-average ROC AUC.
Switched to numpy.interp, which is returning the expected results without errors.
Fixed paths for train and validation dataloaders in the super-resolution notebook.

BoGGoG commented 2 days ago

Thanks! Sorry seems like I missed those.