You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I notice that you set the opt_level='O0', which is FP32 training instead of mixed-precision training. What would happen if using opt_level=O1 or higher opt_level?
The text was updated successfully, but these errors were encountered:
I have tried opt_level=O1 and O2, they gave very close results but didn't show much speedup in training time.
I guess it is due to the communication bottleneck.
I notice that you set the
opt_level='O0'
, which is FP32 training instead of mixed-precision training. What would happen if usingopt_level=O1
or higheropt_level
?The text was updated successfully, but these errors were encountered: