From snuttall35 at rochester.rr.com  Sun Jan 15 03:19:49 2012
From: snuttall35 at rochester.rr.com (snuttall35 at rochester.rr.com)
Date: Sat, 14 Jan 2012 21:19:49 -0500
Subject: [Ifit-users] losing weight and getting tone
Message-ID: <20120115021949.10LR8.172566.root@hrndva-web10-z01>

I have very weak legs.  If I fall or sit on the floor, I have a hard time getting up.  What kind of exercises can I do to improve my les?


From farhi at ill.fr  Tue Feb 14 05:46:19 2012
From: farhi at ill.fr (Emmanuel FARHI)
Date: Tue, 14 Feb 2012 05:46:19 +0100 (CET)
Subject: [Ifit-users] iFit 1.1 now available
Message-ID: <61c1bd8061902629970e8799d201e30d.squirrel@195.83.126.17>

Dear iFit users,

A new release of iFit, version 1.1 has just been pushed on:

                <http://ifit.mccode.org>

You are welcome to get and use it.

The main changes from previous release 1.0.4 are:
* new iData methods (save, diff, hyperbolic functions),
* fix of the Levenberg-Marquardt optimizer,
* improved Error propagation for some unary operators,
* support for import of Analyze MRI (hdr/img) and Crystallographic Binary
File (cbf) format,
* import post-process for HDF RITA2 at PSI files,
* direct drag-drop data files using the open{ext} functions with Matlab >=
2010,
* added many fit models including 2D functions signal=f(x,y).
* Improved speed by e.g. +25%
* a number of minor bug fixes mostly found during the iFit workshop held
at NBI, Copenhagen in January 2012
* improved McStas support (dynamic plot of monitors)

Thanks to all of you for using iFit and giving me the opportunity to
improve it. Do not hesitate to contact me and report potential bugs. I
will certainly do all I can to solve them.

Emmanuel.

-- 
FARHI Emmanuel <farhi at ill.fr>
Groupe DS/CS, ILL4/156, Tel 04 76 20 71 35
ILL, Grenoble



From farhi at ill.fr  Thu Feb 23 15:35:46 2012
From: farhi at ill.fr (Emmanuel FARHI)
Date: Thu, 23 Feb 2012 15:35:46 +0100 (CET)
Subject: [Ifit-users] Problems with iFit Waterfall plot of McStas scans
In-Reply-To: <CAP5ZWsd9RyoHiwqHxDX3RGwoY==7GQ0-sFnB5Mkfzg=WWqK8Ng@mail.gmail.com>
References: <CAP5ZWsd9RyoHiwqHxDX3RGwoY==7GQ0-sFnB5Mkfzg=WWqK8Ng@mail.gmail.com>
Message-ID: <5c384ba2aac1af23bfa19f19823b1101.squirrel@195.83.126.17>

Hello Henrik,

If you import mcstas.sim or dat, only the integral value is recorded
there. The solution is to import all files under the subdirectories, and
search for the monitor you want, then assemble them:

  a=iData('results'); % the full directory...

or, to only get the monitors whou want:

  a=iData({'results/0/PSD_fulllam*','results/1/PSD_fulllam*', .. });

then plot them as waterfall:

  plot(a, 'waterfall')

or even assemble then into a surface, and plot it:

  b = cat(2, a); plot(b);

As I do not know about your data, this may fails, or not show what you
want...

Emmanuel.


On Thu, February 23, 2012 14:50, Henrik Jacobsen wrote:
> Hi Emmanuel
> I have made a scan over one parameter in McStas, and I would like to make
> a
> waterfall plot of this (i.e. plotting intensity vs. wavelength and the
> scan
> parameter for an L_monitor).
>
> I thought this would work by simply using
>
> a=iData('mcstas.sim') (or 'mcstas.dat')
> b=a(14) (I want to plot monitor nr. 14)
> waterfall(b)
>
> But this only plots the integrated intensity of the monitor instead of
> intensity as function af lambda.
> Looking closer at the iData contents, it has all the parameters, MetaData,
> Headers, Parameters and PSD_fulllam_01_ERR, which must be the errorbars on
> my monitor called PSD_fulllam_01.
>
> This is weird, since that monitor is nr. 64 (the last one), while I am
> trying to plot monitor nr. 14.
>
> So it seems there are problems with the loading of the data - perhaps
> because I have many monitors?
>
> I could send you the data, but the folder is 280mb, so I'd rather not do
> that..
>
> Otherwise I'm really enjoying using iFit!
>
> Best
>
> Henrik
>


-- 
FARHI Emmanuel <farhi at ill.fr>
Groupe DS/CS, ILL4/156, Tel 04 76 20 71 35
ILL, Grenoble



From farhi at ill.eu  Fri Mar 16 15:43:51 2012
From: farhi at ill.eu (Emmanuel FARHI)
Date: Fri, 16 Mar 2012 15:43:51 +0100
Subject: [Ifit-users] iFit 1.1.1 is out
Message-ID: <4F6351A7.8090000@ill.eu>

An HTML attachment was scrubbed...
URL: <http://mcstasmailman.esss.dk/pipermail/ifit-users/attachments/20120316/e1384ea0/attachment.html>

From farhi at ill.fr  Fri Mar 16 23:21:15 2012
From: farhi at ill.fr (Emmanuel FARHI)
Date: Fri, 16 Mar 2012 23:21:15 +0100 (CET)
Subject: [Ifit-users] Bugs in iFit
In-Reply-To: <CAP5ZWsfPHyAYWVD70TSW2HuPvL8=oJOndRb2H1oGV4hNLZyh9g@mail.gmail.com>
References: <CAP5ZWsfPHyAYWVD70TSW2HuPvL8=oJOndRb2H1oGV4hNLZyh9g@mail.gmail.com>
Message-ID: <90bbb81c72f42649297b0416d0ef5161.squirrel@195.83.126.17>

Hi Henrik,

Thanks for reporting.

I've fixed this bug, which is related only to McStas simulations started
from iFit.
iFit users launching McStas simulations/scans/optimizations should
download again the archive for the 1.1.1 release. The rest of the package
is not affected.

Sorry for the incovenience.
Emmanuel.

PS: for the display, make sure you indeed use the release 1.1.1. The
display you show corresponds to a previous release. It is now much
cleaner.

On Fri, March 16, 2012 20:45, Henrik Jacobsen wrote:
> Hi Emmanuel
>
> I just downloaded the new version of iFit (yay! looking forward to the
> +12%
> speed), but there is a bug: when running optimizations I get the following
> error: ??? Error using ==> sprintf
>  (more of the output is given in the end of the mail). This does not
> happen
> in the previous version.
>
> I attached a folder with the files that will reproduce the error, by
> running runcmnds.m.
>
> A second minor bug is in the plotting of the output - it is weird with
> text
> in it. I attached a screen shot of it.
>
> But still, I very much enjoy using iFit :)
>
> Have a nice weekend!
>
> Best
>
> Henrik
>
> Finally. Time: 1 [s]
>
> iLoad: Importing file optimize_150m_cold_maxdiv_20/FoM.sim with method
> McStas 1D monitor (looktxt)
> looktxt optimize_150m_cold_maxdiv_20/FoM.sim --fast --binary --headers
> --comment=NULL --silent --metadata=variables --metadata=xlabel
> --metadata=ylabel --metadata=component --metadata=Param --metadata=Creator
> --outfile=/tmp/lk_927612902_B3f3bD
> iData: Setting Signal="Data.N" with length 320 in object id556442 "FoM.sim
> McStas 1D monitor".
> ??? Error using ==> sprintf
> Function is not defined for 'iData' inputs.
>
> Error in ==> mcstas>mcstas_criteria at 702
>       sim_abstract=[ sim_abstract sprintf('\n') '* ' sim(index).Label ...
>
> Error in ==> mcstas>@(pars)mcstas_criteria(pars,options) at 307
>         @(pars) mcstas_criteria(pars, options), pars, options,
> constraints);
>
> Error in ==> fmin_private_wrapper>inline_objective at 613
>     c = feval(fun, pars, varargin{:});         % function=row vector,
> pars=column
>
> Error in ==> fmin_private_wrapper at 208
>     fval = inline_objective(fun, pars, varargin{:});
>
> Error in ==> fminpso at 90
> [pars,fval,exitflag,output] = fmin_private_wrapper(mfilename,
> varargin{:});
>
> Error in ==> mcstas at 306
>       [pars,fval,exitflag,output] = feval(options.optimizer, ...
>
> Error in ==> runcmnds at 226
>             [parameters, monitors, status,
> output]=mcstas('200_elements',p,options);
>


-- 
FARHI Emmanuel <farhi at ill.fr>
Groupe DS/CS, ILL4/156, Tel 04 76 20 71 35
ILL, Grenoble