Daniel@0: function x = gsamp(mu, covar, nsamp)
Daniel@0: %GSAMP	Sample from a Gaussian distribution.
Daniel@0: %
Daniel@0: %	Description
Daniel@0: %
Daniel@0: %	X = GSAMP(MU, COVAR, NSAMP) generates a sample of size NSAMP from a
Daniel@0: %	D-dimensional Gaussian distribution. The Gaussian density has mean
Daniel@0: %	vector MU and covariance matrix COVAR, and the matrix X has NSAMP
Daniel@0: %	rows in which each row represents a D-dimensional sample vector.
Daniel@0: %
Daniel@0: %	See also
Daniel@0: %	GAUSS, DEMGAUSS
Daniel@0: %
Daniel@0: 
Daniel@0: %	Copyright (c) Ian T Nabney (1996-2001)
Daniel@0: 
Daniel@0: d = size(covar, 1);
Daniel@0: 
Daniel@0: mu = reshape(mu, 1, d);   % Ensure that mu is a row vector
Daniel@0: 
Daniel@0: [evec, eval] = eig(covar);
Daniel@0: 
Daniel@0: deig=diag(eval);
Daniel@0: 
Daniel@0: if (~isreal(deig)) | any(deig<0), 
Daniel@0:   warning('Covariance Matrix is not OK, redefined to be positive definite');
Daniel@0:   eval=abs(eval);
Daniel@0: end
Daniel@0: 
Daniel@0: coeffs = randn(nsamp, d)*sqrt(eval);
Daniel@0: 
Daniel@0: x = ones(nsamp, 1)*mu + coeffs*evec';