Hi, the code is mostly in trunk, with one exception. We used JBlas for solving the linear systems in ALS. Unfortunately, this requires certain libs to be installed on the cluster machines, therefore we did not yet commit this to Mahout. Its available on github if anyones interested [1].
I will share the paper once the camera ready version is finished :) [1] https://github.com/sscdotopen/mahout-als/blob/jblas/math/src/main/java/org/apache/mahout/math/als/JBlasAlternatingLeastSquaresSolver.java On 23.07.2013 10:31, Nick Pentreath wrote: > Congrats Sebastian > > Is the code for this already in Mahout trunk? > > Nick > > > On Sun, Jul 21, 2013 at 8:22 PM, Sebastian Schelter <[email protected]> wrote: > >> I'm happy to anounce that a paper called "Distributed Matrix >> Factorization with MapReduce using a series of Broadcast-Joins" written >> by me and my colleagues at TU Berlin has been accepted for publication >> at the ACM Conference on Recommender Systems 2013. >> >> The paper discusses Mahout's latest distributed ALS implementation and >> contains experiments on the Netflix dataset (100M ratings), the Yahoo >> Music dataset (> 700M ratings) and a synthetic dataset generated from >> Netflix with 25M users and more than 5 billion ratings. >> >> Best, >> Sebastian >> >
