Set the multivariate distribution of gausses in this dataset

Question

I need to match the multidimensional Gaussian distribution. We get the average vector and covariance matrix of the nearest multidimensional Gaussian for a given set of audio functions data in python. Sound functions (MFCC coefficients) are an NX 13 matrix, where N is around 4K. Can someone please outline the packages and technique so that they match the gaussian for this data in python?

Answer 1

Use the numpy package. numpy.mean and numpy.cov will give you a gaussian parameter estimate. Assuming you have 13 attributes and N is the number of observations, you need to set rowvar=0 when calling numpy.cov for your matrix N x 13 (or pass the transpose of your matrix as an argument to the function).

If your data is in a numpy data array:

 mean = np.mean(data, axis=0) cov = np.cov(data, rowvar=0)