Zeroed out confusion matrix and AUC = 0 during training when trained with gradable images only

mikevoets / jama16-retina-replication

JAMA 2016; 316(22) Replication Study

https://doi.org/10.1371/journal.pone.0217541

MIT License

110 stars 35 forks source link

Zeroed out confusion matrix and AUC = 0 during training when trained with gradable images only #8

Closed bsabri closed 4 years ago

bsabri commented 5 years ago

I'm observing the same issue reported in #6 when I train with gradable images only. When all eyepacs images are used this does not happen

ZHULLI commented 5 years ago

@bsabri Hello, I am also doing DR, can you tell me how to do it in data preprocessing? I have a dataset but only when the eyepacs.sh is running, the training set is not preprocessed.

mikevoets commented 5 years ago

@bsabri Does this happen every time you train? Or only occasionally?

@ZHULLI What is the exact command you're running? And on what platform?

mikevoets commented 4 years ago

Closing this issue for now because I didn't receive enough information to reproduce this issue.