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
The TensorSpace, LinearSpace and FiniteDimensional classes are pretty bloated, with rather technical methods. It would be very cumbersome having to define all those instances for every custom vectorspace. It shouldn't be necessary either – the definitions seem automatic enough to make it feasible with GHC generics, so a new compound vector space (in particular records, which are basically just direct sums) can be given an empty LinearSpace instance (or directly deriving one, with -XDeriveAnyClass).
The text was updated successfully, but these errors were encountered:
The
TensorSpace
,LinearSpace
andFiniteDimensional
classes are pretty bloated, with rather technical methods. It would be very cumbersome having to define all those instances for every custom vectorspace. It shouldn't be necessary either – the definitions seem automatic enough to make it feasible with GHC generics, so a new compound vector space (in particular records, which are basically just direct sums) can be given an emptyLinearSpace
instance (or directlyderiving
one, with-XDeriveAnyClass
).The text was updated successfully, but these errors were encountered: