A neural gym for training deep learning models to carry out geoscientific image segmentation. Works best with labels generated using https://github.com/Doodleverse/dash_doodler
If TESTTIMEAUG==False, no additional predictions (est_label2, est_label3, est_label4) are made. But code tries to delete those variables anyway. Moved deletion of those predictions into if TESTTIMEAUG == True: condition statement, then delete est_label after E1.append(...).
If
TESTTIMEAUG==False
, no additional predictions (est_label2
,est_label3
,est_label4
) are made. But code tries to delete those variables anyway. Moved deletion of those predictions intoif TESTTIMEAUG == True:
condition statement, then deleteest_label
afterE1.append(...)
.