Can you provide the mdout file that failed here?
On Mon, Feb 1, 2016 at 9:32 AM, Jean-Patrick Francoia <
jeanpatrick.francoia.gmail.com> wrote:
> Cool.
>
> It's buggy however:
>
> Traceback (most recent call last):
> File "/home/djipey/SOFT/amber14/bin/mdout_analyzer.py", line 63, in
> <module>
> mdout = MdoutClass(f)
> File
>
> "/home/djipey/SOFT/amber14/lib/python2.7/site-packages/mdoutanalyzer/mdout.py",
> line 82, in __init__
> if self.properties['imin'] == 5:
> KeyError: 'imin'
>
> Le 01/02/2016 15:05, Jason Swails a écrit :
> > If you're trying to plot data from mdout files, you can always try
> > "mdout_analyzer.py". It's built on top of matplotlib and doesn't require
> > any pre-processing step for mdout files (i.e., you can just type:
> >
> > mdout_analyzer.py mdout1 mdout2 ... mdoutN
> > )
> >
> > HTH,
> > Jason
> >
> > On Mon, Feb 1, 2016 at 8:57 AM, Jean-Patrick Francoia <
> > jeanpatrick.francoia.gmail.com> wrote:
> >
> >> "Plots generally look good"...Yeah maybe with a little formatting as you
> >> suggested. Actually xmgrace has an old look on my platform, a bit like
> >> xleap. That's why I will go for gnuplot.
> >>
> >> I had a bit of spare time, so I wrote a quick script:
> >>
> >> #!/usr/bin/python
> >> # coding: utf-8
> >>
> >> help = """Plot.py
> >>
> >> Script to plot (x, y) data from files
> >>
> >> Usage:
> >> plot.py <files>...
> >>
> >> Options:
> >> -h --help Display help
> >> """
> >>
> >>
> >> from docopt import docopt
> >> import matplotlib.pyplot as plt
> >>
> >> plt.style.use('ggplot')
> >> plt.style.use('seaborn-poster')
> >> plt.tick_params(left='off', bottom='off', top='off', right='off')
> >>
> >>
> >> arguments = docopt(help)
> >>
> >> for arg in arguments['<files>']:
> >> pts_x = []
> >> pts_y = []
> >>
> >> with open(arg, 'r') as f:
> >> for line in f.readlines():
> >> res = [float(value.strip()) for value in line.split(' ')
> >> if value.strip() != '']
> >> if res:
> >> pts_x.append(res[0])
> >> pts_y.append(res[1])
> >>
> >> plt.plot(pts_x, pts_y, label=arg)
> >>
> >> plt.legend()
> >> plt.show()
> >>
> >> Just to try docopt and the new version of matplotlib (which includes
> >> styles). The graph formatting is clearly R-inspired (the devs of
> >> matplotlib finally underestood...).
> >>
> >> Usage:
> >> ./plot.py summary.EPTOT summary.ETOT
> >>
> >> Result:
> >> http://i.imgur.com/FPdBfsz.png
> >>
> >> Le 01/02/2016 14:08, David A Case a écrit :
> >>> On Mon, Feb 01, 2016, Lorenzo Gontrani wrote:
> >>>
> >>>> xmgrace is great..and lightweight.
> >>> Just some (off-topic) comments about why I like xmgrace:
> >>>
> >>> 1. Plots generally look good and can be interactively edited if they
> >> don't.
> >>> Consider modifying your ~/.grace/templates/Default.agr file to specify
> >>> Helvetica as "font 0" (the default) and change 'xaxis label char size'
> >> to 1.25
> >>> (from 1.0); same for yaxis.
> >>>
> >>> 2. A single .agr file has both the data and the formatting in a single
> >> file.
> >>> This is good if you come back months or years later and don't remember
> >> your
> >>> file naming conventions.
> >>>
> >>> 3. You can also write simple scripts to systematically modify lots of
> >>> files, to replace the data but keep the formatting, etc. Once you grok
> >> the
> >>> file syntax (which is pretty straightforward), it's about as easy to
> >> write
> >>> scripts to create or manipulate xmgrace files as it is to write gnuplot
> >> or
> >>> matplotlib scripts.
> >>>
> >>> 4. Consider aliasing xmgrace to 'xmgrace -hdevice PDF'. This means
> that
> >>> "print" will create a PDF file (rather than the default PostScript).
> >> This is
> >>> likely to be what most people want.
> >>>
> >>> ....dac
> >>>
> >>> (OK: I realize I've probably started a flame war here... Next up: why
> you
> >>> should write all your code in LISP.)
> >>>
> >>> _______________________________________________
> >>> AMBER mailing list
> >>> AMBER.ambermd.org
> >>> http://lists.ambermd.org/mailman/listinfo/amber
> >>
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber
> >>
> >
> >
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 01 2016 - 07:00:05 PST
