Main Page

From Eigen
Revision as of 16:45, 12 January 2009 by Bjacob (Talk | contribs)

Jump to: navigation, search

The new Eigen users forum just opened!

If you are still interested in the old version, Eigen 1, please bookmark it as this link is going to be removed from here.

Overview

Eigen 2 is a C++ template library for linear algebra: vectors, matrices, and related algorithms. It is:

  • Versatile. (See modules and tutorial). Eigen handles, without code duplication, and in a completely integrated way:
    • both fixed-size and dynamic-size matrices and vectors.
    • both dense and sparse (the latter is still experimental) matrices and vectors.
    • both plain matrices/vectors and abstract expressions.
    • both column-major (the default) and row-major matrix storage.
    • both basic matrix/vector manipulation and many more advanced, specialized modules providing algorithms for linear algebra, geometry, quaternions, or advanced array manipulation.
  • Fast. (See benchmark).
    • Expression templates allow to intelligently remove temporaries and enable lazy evaluation, when that is appropriate -- Eigen takes care of this automatically and handles aliasing too in most cases.
    • Explicit vectorization is performed for the SSE (2 and later) and AltiVec instruction sets, with graceful fallback to non-vectorized code. Expression templates allow to perform these optimizations globally for whole expressions.
    • With fixed-size objects, dynamic memory allocation is avoided, and the loops are unrolled when that makes sense.
    • For large matrices, special attention is paid to cache-friendliness.
  • Elegant. (See API showcase). The API is extremely clean and expressive, thanks to expression templates. Implementing an algorithm on top of Eigen feels like just copying pseudocode. You can use complex expressions and still rely on Eigen to produce optimized code: there is no need for you to manually decompose expressions into small steps.
  • Compiler-friendy. Eigen has very reasonable compilation times at least with GCC, compared to other C++ libraries based on expression templates and heavy metaprogramming. Eigen is also standard C++ and supports various compilers.

FAQ

Frequently asked questions are here.

Documentation

The documentation is here.

It includes a Tutorial.

To learn about the internals of Eigen, read this example and check out this page about some of Eigen's internal mechanisms: EigenInternals.

Requirements

Eigen 2 doesn't have any dependency. It just uses a little the C++ standard library.

It uses the CMake build system. However, this is only to build the documentation and unit-tests, and to automate installation. If you just want to use Eigen, you can use the header files right away. There is no binary library to link to (pure template library), and no configured header file.

Download

Here is the source tarball for the latest release: eigen-2.0-beta6.tar.bz2

Alternatively, you can checkout the development tree by anonymous svn, by doing:

svn co svn://anonsvn.kde.org/home/kde/trunk/kdesupport/eigen2

or view it online here

License

Eigen is Free Software. It is licensed under the LGPL3+. As an alternative license choice, Eigen is also licensed under the GPL2+.

Virtually any software may use Eigen. Even closed-source software may use Eigen without having to disclose its own source code.

See the Licensing FAQ.

Credits

Core developers (in alphabetical order):

  • Gaël Guennebaud
  • Benoît Jacob

Contributors (in alphabetical order):

  • David Benjamin (the owls)
  • Armin Berres (MSVC compatibility fixes, GCC warning fixes)
  • Daniel Gómez (several improvements, especially in the Sparse module)
  • Konstantinos Margaritis (Altivec vectorization)
  • Christian Mayer (early code review and input in technical/design discussions)
  • Michael Olbrich (initial meta loop unroller and early benchmarking)
  • Kenneth Riddile (MSVC compatibility fixes, exception throwing)

Special thanks to Tuxfamily for the wonderful quality of their services!

Compiler support

Eigen is standard C++98 and so should theoretically be compatible with any compliant compiler. Of course, in practice, things are slightly different.

Eigen is being successfully used with the following compilers:

  • GCC, version 3.3 and newer. Very good performance with GCC 4.2 and newer.
  • MSVC (Visual Studio), 2005 and newer. Vectorization is enabled with 2008 and newer.
  • ICC, recent versions. Very good performance.
  • MinGW, recent versions. Performance is poor because MinGW uses GCC 3. This problem will go away whenever MinGW upgrades to GCC 4.

Here are some comments about GCC compiler flags.

  • At least some optimization is mandatory to get even remotely decent speed. -O1 gives something decent for a debug mode, at 30-60% of the optimal speed. -O2 generally gives optimal speed. -O3 does not have much advantages over -O2, in our experience.
  • Debugging info with -g (equivalently -g2) can increase dramatically the executable file's size. This is always the case, but even more so with Eigen.
  • Disabling asserts, by defining -DNDEBUG or -DEIGEN_NO_DEBUG, improves performance in some cases.
  • Vectorization is automatically enabled if a SIMD instruction set is enabled by the compiler. On the x86 platform, SSE2 is not enabled by default and you need to pass the -msse2 option.

Projects using Eigen 2

While Eigen 2 is still a very young library, even before its official release it is already being used by a number of applications and projects:

  • Various KDE related projects such as some screensavers, kgllib, kglengine2d, solidkreator, etc.
  • Koffice2 (KDE's office suite), in particular Krita, the painting and image editing module. Eigen is also used a bit by KSpread, the spreadsheet module, for matrix functions such as MINVERSE, MMULT, MDETERM.
  • Avogadro, an opensource advanced molecular editor.
  • VcgLib, an opensource C++ template library for the manipulation and processing of triangle and tetrahedral meshes. (switched from home made math classes)
  • MeshLab, an opensource software for the processing and editing of unstructured 3D triangular meshes and point cloud. (switched from vcglib's math classes)
  • Expe, an experimental framework for the rapid prototyping of graphics applications. No release yet, but it uses 90% of Eigen's features. (switched from home made math classes).
  • libmv, an opensource structure from motion library. (switched from FLENS)
  • The Yujin Robot company uses Eigen for the navigation and arm control of their next gen robots. (switched from blitz, ublas and tvmet)
  • The Robotic Operating System (ROS) developed by Willow Garage.

If you are aware of some interesting projects using Eigen, please send us a message or directly edit this wiki page !

Get support

Need help using Eigen? Try this:

  • The users forum is your best resource.
  • Our IRC channel is #eigen on irc.freenode.net.
  • Want to discuss something with the developers? Think you've found a bug? Use our mailing list.

Mailing list

Our mailing list is the central point for discussion of Eigen development, bugs, feature requests, etc.

  • To subscribe, send a mail with subject "subscribe" to eigen-request at lists tuxfamily org.
  • To unsubscribe, send a mail with subject "unsubscribe" to eigen-request at lists tuxfamily org.

Once you are subscribed, you may post to eigen at lists tuxfamily org.

You can also browse the archive

You can also contact us by IRC : #eigen on irc.freenode.net.

For ordinary support questions, please see the Get support section instead.