While calculating AUC , the abnormal pixel are filtered with mask value of 0 , while in the agrivision data the anomaly pixels are masked with value of 1, am I missing something

[sambalshikhar]

ASD/metrics/auroc.py

Line 38 in a00d10f

anomaly_scores = amap[np.where(gt == 0)]

[Jingtao-Li-CVer]

Sorry for the late reply. It is indeed a little confusing. I check the code and the anomaly pixels in agrivision data are actually masked with 0. I am pleased to discuss it with you further about this problem