Closed davidbp closed 5 years ago
Yes, the corresponding method for Adjoint
s is missing, but the one for Transpose
s exists:
julia> Wt = copy(transpose(W));
julia> @btime Wt*x_sp;
501.161 ns (1 allocation: 96 bytes)
julia> @btime transpose(W)*x_sp;
462.653 ns (3 allocations: 128 bytes)
I started translating the transpose method to adjoints, PR is coming soonish...
When multiplying Dense matrices by sparse vectors the speed seems to degrade too much if the dense matrix is an adjoint.
This seems to be related with #32195 but this is not matrix-matrix multiply. This is matrix-vector multiply