This problem keeps coming up for me over the years, and there must be a standard solution, so perhaps someone knows it, I can't ever google it up.
Say I have an array A of any number of dimensions, and I want to apply an indexing operation to the last dimension. Now, if it's 2D, I just use:
If it's 3D, I use:
but what if it's n-D? Afaik there is no syntax in the language for this, so I'd have to use some function. A function "index_into()", for instance, to give:
B = index_into(A, n, ii);
At the moment, I always end up doing some horrible mash like:
X = permute(A, [ndims(A) ...
X = X(ii, :);
X = reshape(X, ...
B = permute(X, ...
Surely there must be a better way...? Any suggestions?