Fixed mapping for computing a square Euclidean distance matrix
D = DISTM(A,B)
Computation of the distance matrix D between two sets A and B. If A has M objects and B has N objects, then D is [M x N]. If A and B are datasets, then D is a dataset as well with the labels defined by the labels of A and the feature labels defined by the labels of B.
Unlabeled objects in B are neglected, unless B is entirely unlabeled. If A is not a dataset, but a matrix of doubles then D is also not a dataset, but a set of doubles.
NOTE (1) DISTM was in older Matlab versions for large datasets considerable faster than PDIST2. Now it uses internally PDIST2.
NOTE (2) The default distance meansure used by DISTM is the square of the Euclidean distance. This can be changed by calling DISTYPE which sets a global used by DISTM.
NOTE (3) DISTM(A,B) is equivalent to A*PROXM(B,'d',2)).