RangerAdaBelief setstate
soloice opened this issue · comments
Encountered some problem when I tried to resume previous training with RangerAdaBelief optimizer.
Should this line be RangerAdaBelief instead of Ranger?
That's correct, thanks for pointing out! Did not test resuming. I'll correct it in the next release
Update it in ranger-adabelief==0.1.0 (available on pip now). See this line https://github.com/juntang-zhuang/Adabelief-Optimizer/blob/update_0.1.0/pypi_packages/ranger_adabelief_0.1.0/ranger_adabelief/ranger_adabelief.py#L101