Re: [ESPResSo-devel] Correlators

[Ulf Schiller]

On 12/06/2011 05:31 PM, Stefan Kesselheim wrote:
> Hi,
>
> Am 06.12.2011 um 17:01 schrieb Ulf Schiller:
>
> > Yes, the core implementation was probably not submitted as it was just a 
> > byproduct of testing fluctuation dissipation relations and Green-Kubo stuff. 
> > For some reason I was thinking it was used in one of the test scripts, but I 
> > just verified it is not. Apparently there is no test script for fluctuation 
> > dissipation in the repository. Maybe someone wants to try...?
>
> Yes, definitely :-). I never checked it explicitly. You once gave me a copy of 
> your Espresso-LB with boundary collisions and 2nd order Langevin and so on, but 
> w rsp. to acf there was nothing but stubs :-).

Well, I eventually found it more convenient to use external tools for 
the data analysis. Like Numpy, e.g.

def autocorr(x):
    result = numpy.correlate(x, x, mode='full')
    return result[result.size/2:]

Intervals and lag times could be applied through appropriate mapping 
functions. That's something to think about when the Python interface 
arrives ;-)

Cheers,
Ulf

smime.p7s
Description: S/MIME Cryptographic Signature