weighted GCCA about cca_zoo HOT 3 CLOSED

Thanks for this idea. I've gone ahead and implemented it in the latest version 1.1.21 (both weighted views and observed/unobserved). For the moment I've just got the exact solution (i.e. directly changing the eigenvalue problem). I should be able to improve efficiency in the future by making a version that uses alternating least squares solvers + need to update the docs so will leave this open.

from cca_zoo.

This particular case should now work by using GCCA().fit(view1,view2,...,view_n, view_weights=view_weights, K=K) where view_weights is an array with one weight for each view and K is an array with (n,N) where n is the number of views and N is the total number of samples and its entries are 1 or 0 depending if that sample is observed for that view

from cca_zoo.

small edit - slightly changed the way methods are built and constructed so the above is now GCCA(view_weights=view_weights, K=K).fit(view1,view2,...,view_n)

from cca_zoo.