A new release (July 10, 2012) of GraphLab!
From the webpage:
Overview
Designing and implementing efficient and provably correct parallel machine learning (ML) algorithms can be very challenging. Existing high-level parallel abstractions like MapReduce are often insufficiently expressive while low-level tools like MPI and Pthreads leave ML experts repeatedly solving the same design challenges. By targeting common patterns in ML, we developed GraphLab, which improves upon abstractions like MapReduce by compactly expressing asynchronous iterative algorithms with sparse computational dependencies while ensuring data consistency and achieving a high degree of parallel performance.
The new GraphLab 2.1 features:
- a new GraphLab 2 abstraction
- Fully Distributed with HDFS integration
- New toolkits
- Collaborative Filtering
- Clustering
- Text Modeling
- Computer Vision
- Improved Build system and documentation
Go to http://graphlab.org for details.
If you want to get started with GraphLab today download the source or clone us Google code. We recommend cloning to get the latest features and bug fixes.
hg clone https://code.google.com/p/graphlabapi/If you don't have mercurial (hg) you can get it from http://mercurial.selenic.com/.
I almost didn’t find the download link. Just larger than anything else on the page, white letters, black backgroud, at: http:www.graphlab.org. Kept looking for a drop down menu item, etc.
Shows even the clearest presentation can be missed by a user. 😉
Now to get this puppy running on my local box.