Re: [cclib-users] A question about LPA

[Karol M. L. <kar...@gm...>]

Hi,

The differences are just due to me using an ultra-stable isntallation
on this particular computer. I will test with newer version on
another machine (probably tommorrow).

- Karol

On Jan 27 2012, Martin Rahm wrote:
> Hi Karol,
> 
> Interesting. I'm using Python 2.7.1, so different but newer than yours,
> with GCC 4.2.1, if that matters. cclib is 1.01 (the latest), and finally,
> Numpy 1.5.1, which also is newer than yours.
> 
> It's all on a OS X 10.6.8 machine.
> 
> Clearly there are some external differences, however, going back to older
> versions are cumbersome. Does this mean one always have to use older
> packages to be sure, or is there a possibility to make cclib work with the
> new ones?
> 
> Thanks again for your hard work & with best regards,
> Martin
> 
> On Fri, Jan 27, 2012 at 4:51 PM, Karol M. Langner
> <kar...@gm...>wrote:
> 
> > Hi Martin,
> >
> > I've come around to this issue. First of all, in this case I don't get
> > the complex number error your report, and my fragcharges sum up nicely.
> > That is a good thing, because it means the cause is perhaps external
> > to cclib.
> >
> > I'm using Python 2.6 here, but I assume that would not be a problem.
> > My NumPy version is 1.4.1 and I'm using the latest cclib revision from svn.
> > What are your version?
> >
> > - Karol
> >
> > On Jan 12 2012, Martin Rahm wrote:
> > > Hello again,
> > >
> > > I've come across another case of the "complex number" error. This time
> > in a
> > > very small diatomic.
> > >
> > > Running LPA on a M062X/cc-pwCV5Z wave function of ClF, and gets this
> > error:
> > >
> > > [LPA <cclib.parser.data.ccData object at 0x1015ba1d0> INFO] Creating
> > > attribute aoresults: [array[2]]
> > >
> > /Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/cclib/method/lpa.py:99:
> > > ComplexWarning: Casting complex values to real discards the imaginary
> > part
> > >   self.aoresults[spin][i] = numpy.multiply(temp1, temp2).astype("d")
> > >
> > > This is quite cumbersome, since there is really no way of breaking
> > symmetry
> > > of a linear molecule (which solved the problem for larger molecules)!
> > >
> > > I've attached my log file, and also my script just in case.
> > >
> > > I hope you have a quick fix :)
> > >
> > > Best regards,
> > > Martin
> >
> > > Archive:  /tmp/ClF-M06-cc-pwCV5Z.log.zip
> > >   Length      Date    Time    Name
> > > ---------  ---------- -----   ----
> > >   8430472  2012-01-11 09:10   ClF-M06-cc-pwCV5Z.log
> > > ---------                     -------
> > >   8430472                     1 file
> >
> >
> > >
> > ------------------------------------------------------------------------------
> > > RSA(R) Conference 2012
> > > Mar 27 - Feb 2
> > > Save $400 by Jan. 27
> > > Register now!
> > > http://p.sf.net/sfu/rsa-sfdev2dev2
> >
> > > _______________________________________________
> > > cclib-users mailing list
> > > ccl...@li...
> > > https://lists.sourceforge.net/lists/listinfo/cclib-users
> >
> >
> > --
> > written by Karol Langner
> > Sat Jan 28 01:46:35 CET 2012
> >

-- 
written by Karol Langner
Sat Jan 28 03:40:06 CET 2012