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
This is left out deliberately as it doesn't make sense unless we have a dense format. A COO with arange is strictly worse than a NumPy array with arange.
Activity
prady0t commentedon Aug 13, 2026
Sure! Let's implement this.
hameerabbasi commentedon Aug 13, 2026
This is left out deliberately as it doesn't make sense unless we have a dense format. A COO with arange is strictly worse than a NumPy array with
arange.lucascolley commentedon Aug 13, 2026
Does finch-tensor plan to have a dense format?
hameerabbasi commentedon Aug 13, 2026
It does, yes.
hameerabbasi commentedon Aug 13, 2026
We actually accept NumPy arrays in PyData/Sparse where it makes sense for us.