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
It would be good to document these numbers to better understand why these specific values have been chosen.
Also, these values will probably not be appropriate if real_type is not chosen to be double floating precision. Perhaps a better idea would be to define EPSILON and EPSMAC in terms of std::limits::epsilon().
The text was updated successfully, but these errors were encountered:
Givens rotation in (F)GMRES solvers is implemented as
where
EPSILON
andEPSMAC
are defined asIt would be good to document these numbers to better understand why these specific values have been chosen.
Also, these values will probably not be appropriate if
real_type
is not chosen to be double floating precision. Perhaps a better idea would be to defineEPSILON
andEPSMAC
in terms ofstd::limits::epsilon()
.The text was updated successfully, but these errors were encountered: