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
Im currently getting started with your package (very nice work btw!) and I am wondering if there is a way to make something like this :
using SuiteSparseGraphsBLAS, Graphs
A_T =GBMatrix{Bool}((adjacency_matrix(graph, Bool; dir=:in))) # Some sparse GB_Matrix of Bool
out_degree =reduce(+, A_T; dims=2) # count the number of true elements on each line
without having this error :
ArgumentError: The SuiteSparse:GraphBLAS reduce function only supports monoids where T x T -> T.
Please pass a function whose output type matches both input types.
The way I understand it, there is no type promotion happening in the reduce function. Since using the boolean matrix is quite faster for some operations, I'd be nice to have a way of doing this. Do you know any way around this ? Is this something that could be made possible at some point ?
Thank you for this nice package!
The text was updated successfully, but these errors were encountered:
Hello,
Im currently getting started with your package (very nice work btw!) and I am wondering if there is a way to make something like this :
without having this error :
The way I understand it, there is no type promotion happening in the reduce function. Since using the boolean matrix is quite faster for some operations, I'd be nice to have a way of doing this. Do you know any way around this ? Is this something that could be made possible at some point ?
Thank you for this nice package!
The text was updated successfully, but these errors were encountered: