temperature_scaling icon indicating copy to clipboard operation
temperature_scaling copied to clipboard

Published 20 hours ago • verified publisher icon gpleiss

model.eval for temperature scaling?

Open vivkul opened this issue 5 years ago • 1 comments

Hi, I was wondering that as dropout is used, shouldn't we be using original_model.eval() for temperature scaling?

As the logits obtained using one forward pass is used to get temperature (instead of a different forward pass for each LBFGS step). Thanks!

vivkul avatar May 18 '19 11:05 vivkul

Wondering the same thing. The code might not use dropout as it were set to 0, but it obviously use batchnorm.

Any answer on this yet?

pvmilk avatar May 08 '20 05:05 pvmilk