We upgraded our Sourceforge repository to the newest version as recommended by Sourceforge. The new path to Shark is now (a Shark directory will be created as a subfolder – if you want the tree contents directly in the current directory, add a space and period . to the end of the command):
Application examples
Contributing to Shark
Show page sourcesvn co https://svn.code.sf.net/p/shark-project/code/trunk/Shark
We are happy to announce that the new version 3.0 of Shark has now moved to the trunk of our svn repository, replacing the last stable version of Shark 2.0.
Shark’s high-level interface is regarded as stable. But until a group of lower-level changes is included, Shark 3.0 is considered in beta stage. Note that Shark 3.0 is already actively used for everyday research by several machine learning groups.
Also note that some tutorials still need to be updated to reflect recent interface changes, so code from some tutorials may not compile. Especially the Data tutorials are known to be outdated.
Feel free to try and test Shark, we are happy about any feedback!
You can download Shark from our svn repository:
svn co https://svn.code.sf.net/p/shark-project/code/trunk/Shark
There is currently a known problem for MacOs users that they cannot compile using gcc 4.2.1. In this case you have to update to a newer version of the gcc or use clang as described in our faq.
October 22, 2011: We are happy to announce that our alpha-release of Shark 3.0 has won the Gold Prize at this year’s Open Source Software World Challenge 2011.
September 18 2011: The first pre-release version of Shark has been finalized and uploaded on occasion of the Open Source Software World Challenge 2011. We thank the organizers, hosts, and sponsors of the competition for the opportunity to participate in such a challenge.
We for convenience again provide the Shark project development plan as submitted to the first (registration) stage of the competition.
We also provide direct entry points to all deliverables for the competition:
Source code: please find a complete source code package on the main Shark download page. There, you will also find pre-compiled binary packages for all major platforms, and other download links.
License information: please find all license information on the Credits and Copyright page.
Installation instructions: Shark can be installed either using pre-compiled binary packages or by compiling from source code. Some attention has to be paid to installation/availability of the Boost C++ library as an external dependency. The Shark installation page will cover all aspects of the installation in detail.
We recognize that installation practices, systems, and experience may vary between different companies, labs, and universities. If there are any gaps or ambiguities in the installation instructions, or documentation in general, please do not hesitate to point them out to us and we will fix any remaining issues immediately.
Other documentation: We provide class and member documentation extracted from in-code documentation, as well as a vast number of tutorials, test sets, and examples. Please look at our guide to the different entry points into these documentation options here.
Introductory videos: As part of the submission, we created a video giving a short overview over the library:
We further provide an introductory video on Shark installation under the Windows operating system on the Shark installation page.
