Registered Member
|
hi,
i'd like to solve this Problem: vector = matrix * matrix^-1 * vector hs size = (ns x ns), hf size = (nf x ns), u size (ns x 1) uf = hf * hs^-1*u my first solution was uf= Hf*(Hs.fullPivLu().solve(u)) but i need the product of hf*hs^-1 like kopplungsmatrix = Hf*Hs.inverse(); uf = kopplungsmatrix*u; can i use another solver like this? Greedings Ajax |
Moderator
|
In the future you will be able to trigger a solving on the right by writing:
kopplungsmatrix = Hf * Hs.lu().inverse(); In the meantime you have to workaround: kopplungsmatrix = Hs.transpose().lu().solve(H.transpose()); kopplungsmatrix.transposeInPlace(); |
Registered Member
|
Registered users: Baidu [Spider], Bing [Bot], Google [Bot], rblackwell