Matrix is not positive definite

[mschart]

Hi there,

Thanks for this nice toolbox. I tried the synthetic.py with time-binned neural data (1000 observations, 124 neurons). I.e. it was not binary, and m.resample_model() runs for a couple of iterations until it says

LinAlgError: Matrix is not positive definite

Maybe you can tell me what's wrong?

Also, how could one run this in parallel ("UserWarning: PyPolyaGamma was not installed with OpenMP. Calls to 'pgdrawvpar' will be replaced with a serial implementation.")

Cheers,

Michael