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
@ChasingNeutrons got some NaNs in the standard deviation of the out-scatter transport cross section calculation when generating cross sections for a reactor problem. We're still not sure of where they come from, so needs checking!
The text was updated successfully, but these errors were encountered:
I think that gfortran can take the -ffpe-trap=invalid flag and then it will crash if the NaN appears (displaying a backtrace). Maybe it will be helpful to find the origin of the NaNs?
@ChasingNeutrons got some NaNs in the standard deviation of the out-scatter transport cross section calculation when generating cross sections for a reactor problem. We're still not sure of where they come from, so needs checking!
The text was updated successfully, but these errors were encountered: