During the calculation of a Gaussian Mixture Model I have to calculate the pdf() of the multivariate Gaussian distribution. I do this using the scipy multivariate_normal() method. Now during the calculations I encounter the situation that I get an error which says
LinAlgError: singular matrix
Setting the allow_singular parameter of the scipy multivariate_normal() = True
circumvents this error and I get a quite meaningful result:
So how is scipy "allowing" for singularity and what are the consequences for the calculations? --> If there are no consequences, strictly speaking, this parameter could be set to TRUE by default (but it isn't).