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
Currently PGM requires the losses to be smooth (f.is_smooth = True) in order to run although at the same time some of our losses are not smooth, for example, PoissonLoss.
It is still often possible to run PGM (see Poisson examples and TV example) even though the losses are not smooth everywhere by forcing the is_smooth flag to be true.
I suggest re-thinking the use of is_smooth in conjunction with PGM.
Currently PGM requires the losses to be smooth (f.is_smooth = True) in order to run although at the same time some of our losses are not smooth, for example, PoissonLoss.
It is still often possible to run PGM (see Poisson examples and TV example) even though the losses are not smooth everywhere by forcing the is_smooth flag to be true.
I suggest re-thinking the use of is_smooth in conjunction with PGM.
Originally posted by @tbalke in #78 (comment)
The text was updated successfully, but these errors were encountered: