Ben Erichson

I guess cuSOLVER won't replace CULA very soon. So far the methods are seeming to be very limited. Just to mention SVD, here only the 'full' SVD decomposition is available,...

Chris, I cannot follow you. Why do you expect T, the principal components, to have orthonormal columns? PCA imposes only constraints on the weight vectors (I guess that is what...

I like this idea a lot. It should make sense for most of current algorithms.

@tsaigaga thanks for spotting this. Do you like to push the fix to the package? Best, ben

Hello Constantin, The rsvd package assumes that the input matrix has low-rank structure. In the example above, you are generating a full-rank matrix. I don't expect rsvd to work in...