James Chapman

Ah I see - yes only the deep models. I'd be curious which model is running slow. In the alternating optimisation methods the bottleneck will be scikitlearn regression solvers. In...

Also worth saying if you use the Deep CCA methods with single layer linear encoders they should converge to CCA and then could use GPU via that route (if you...

Ah interesting! Must admit my implementation of NCCA is definitely functional rather than optimized. If there's a faster nearest neighbour algo that can be imported I'd defo drop it in....

https://towardsdatascience.com/make-knn-300-times-faster-than-scikit-learns-in-20-lines-5e29d74e76bb Might take a look at this

Just seen the edit! that's cool!

This is really nice work! Sorry for my slowness getting back to you. Had to properly read the MATLAB code to see how it works. Let me know your thoughts...

A particularly fiddly bit is that the permutations are on the U and V matrices which are subspaces of X and Y.

My guess is that this is a situation where the number of samples==number of features in one/both views? If yes (this was the situation where I was able to reproduce...

So the gist of the way the solver works is to get the PCA components of the original data and runs the algorithm on the reduced data. This is mathematically...

Some examples of the sort of thing in the tutorial notebooks. Things like plotting the projections/scores or visualising the model weights.