frankkramer-lab / aucmedi

a framework for Automated Classification of Medical Images
https://frankkramer-lab.github.io/aucmedi/
GNU General Public License v3.0
38 stars 12 forks source link

Training freeze at end of first epoch (validation computation) #198

Open muellerdo opened 1 year ago

muellerdo commented 1 year ago

https://github.com/keras-team/keras/issues/8595

Deathlymad commented 1 year ago

mutiple people in the referenced that setting validation steps solved it. Would it be a reasonable fix to enforce this in model.py?

Some people also referenced locks due to using multiple GPUs but i don't think there is a reasonable fix for this.