a simple library to analyze data (with McCode and Phonons/DFT hooks)
Matlab Fortran C HTML Python Mathematica Other
Clone or download
Latest commit c7febb9 Aug 10, 2018
Failed to load latest commit information.
Libraries Loaders: read_gcode also partition the GCode objets and display when … Aug 9, 2018
Objects Close #146 Aug 10, 2018
Tests iData_Sqw2D: added more tests Aug 7, 2018
COPYING update of COPYING files Aug 10, 2016
README.md Docs: renamed Treatment -> Neutron Scattering. Close #137. Oct 10, 2017
mkdist standalone: include Python iFit as a launcher. Aug 9, 2017


Welcome to the iFit/iData package

                      E. Farhi, ILL/CS [farhi@ill.fr]
                      Version @IFIT_VERSION@ - @IFIT_DATE@


This library aims at providing basic functionality to achieve some of the general tasks needed for scientific data analysis: Load, Plot, Save, Fit, Math operations, define and use Models

It also includes specific 'applications' for neutron scattering:


Requirements for standalone (binary) package: NONE Stand-alone versions do not require a Matlab license, and have no dependency except the Matlab Compiler Runtime (MCR). This latter is included in some of the packages (Mac OSX), or has to be installed separately (Windows, Linux). You can get the MCR installer at http://ifit.mccode.org/Downloads/binary

Under MacOSX, you may need to install Java 6. Get it at: http://ifit.mccode.org/Downloads/binary/mac64/javaforosx.dmg

Requirements for Source package Matlab (any version from 7.x, R2007a), possibly a C compiler for the looktxt
and the cbf_uncompress MeX.


Installation from Source package Copy the library directories where-ever you want or in MALTAB/toolbox/local: /home/joe/Matlab/iFit or /opt/MATLAB/R2010a/toolbox/local/iFit

Then start Matlab and type in, e.g.:

>> addpath(genpath('/home/joe/Matlab/iFit'))


>> addpath(genpath('/opt/MATLAB/R2010a/toolbox/local/iFit'))

Installation for standalone (binary) package Install the MCR (see above, prefer /opt/MATLAB location on Linux systems), then extract the iFit binary package and launch 'ifit'.

MacOSX Drag and drop the iFit icon into your Applications folder. To open the standalone App the first time, use Ctrl-Click on the icon.

If you get an error dialogue stating you need Java 6, install it from: http://ifit.mccode.org/Downloads/binary/mac64/javaforosx.dmg

Quick start type at Matlab prompt:

>> addpath(genpath('/path/to/iFit'))
>> doc(iData)

Then refer to the Quick Start tutorial (in iFit/Docs/QuickStart.html).

The miFit interface will be automaticaly opened. It allows to drag-and-drop data files, plot them, do basic data treatment, fitting, export, ...

Useful links


You can register to the iFit mailing list at http://mail.mccode.org/cgi-bin/mailman/listinfo/ifit-users Send messages to the ifit-users@mail.mccode.org. Help pages are available at http://ifit.mccode.org

License: EUPL

Basically this is open-source. Use it if you find it useful, and enrich it. If you do produce new methods, please send them back to me so that they are added in the software and thus benefit to the community.

In short, you can use, copy, distribute and modify the Code. However, a number of restrictions apply, especially when producing derived work (that is modify and redistribute the code in other products). In particular, the derived work must be licensed under the EUPL or a Compatible License, label all modifications explicitly, distribute the Source Code, and cite the Original work. The Source code of iFit is freely available at http://ifit.mccode.org

A number of additions, included in the software, where obtained from the Matlab Central contributions, and are BSD licensed.

Matlab is a registered trademark of The Mathworks Inc.

Contributions are listed in the Credits page at http://ifit.mccode.org/Credits.html


This is not a professional tool, and there is no Dev team to actively take care of it. Expect occasional failures and bugs. However, I try my best to make the software efficient and reliable.

$Revision$ - $Date$