circ_vmrnd

[philippberens]

I found a small inconsistency in the circ_vmrnd program. On line 50, it should read:

alpha = 2pirand(n,1)-pi;

Currently the "-pi" isn't in the code, which means for large values of kappa, the distribution takes values on the interval (-pi,pi), while for kappa close to 0 it takes values between (0,2pi).

[jCalderTravis]

In this case of small kappa, the code also does not reshape the output. In contrast, usually when the input argument, n, is a two-element vector the output is reshaped. Shall I change this and submit a pull request?

Thank you for an amazing toolbox.