| clear query|facets|time |
Search criteria: .
Results from 61 to 70 from
697 (2.172s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: Discussion Of ML environment/MR, Mahout - Mahout - [mail # dev]
|
|
...Only partially. There are tools to produce the interaction data into training and test set, and to measure the RMSE on the test set. However, there is no tooling for cross-validation a...
|
|
|
Author: Sebastian Schelter,
2013-03-19, 16:11
|
|
|
Re: Discussion Of ML environment/MR, Mahout - Mahout - [mail # dev]
|
|
...Played a little more with the code, it works astonishingly well. I was totally off in my expectations. I was able to run an iteration of ALS (two map-only jobs) on the Yahoo Songs data...
|
|
|
Author: Sebastian Schelter,
2013-03-19, 15:09
|
|
|
Re: ALS-WR on Million Song dataset - Mahout - [mail # user]
|
|
...Hi JU, We recently rewrote the factorization code, it should be much faster now. You should use the current trunk, make Hadoop schedule only one mapper per machine (with -Dmapred.taskt...
|
|
|
Author: Sebastian Schelter,
2013-03-19, 15:07
|
|
|
Re: Question for Collaborative Filtering with ALS-WR - Mahout - [mail # user]
|
|
...Hi Qi, its absolutely possible. Have a look at PredictRatingsMapper in org.apache.mahout.cf.taste.hadoop.als.FactorizationEvaluator This mapper loads the feature matrices U and M...
|
|
|
Author: Sebastian Schelter,
2013-03-19, 11:55
|
|
|
Re: ALS-WR on Million Song dataset - Mahout - [mail # user]
|
|
...You should also be aware that the alpha parameter comes from a formula the authors introduce to measure the "confidence" in the observed values: confidence = 1 + alpha * observed_value...
|
|
|
Author: Sebastian Schelter,
2013-03-18, 17:12
|
|
|
Re: ALS-WR on Million Song dataset - Mahout - [mail # user]
|
|
...JU, are you refering to this dataset? http://labrosa.ee.columbia.edu/millionsong/tasteprofile On 18.03.2013 17:47, Sean Owen wrote:...
|
|
|
Author: Sebastian Schelter,
2013-03-18, 16:52
|
|
|
[MAHOUT-1166] Multithreaded version of distributed ALS - Mahout - [issue]
|
|
...Our implementation of ALS broadcasts the feature matrices in each iteration. Therefore, it makes sense to run the mappers in multithreaded mode to not have to load one copy of the feature ma...
|
|
|
http://issues.apache.org/jira/browse/MAHOUT-1166
Author: Sebastian Schelter,
2013-03-18, 13:32
|
|
|
Re: reproducibility - Mahout - [mail # user]
|
|
...These also have a stochastic element, as the Hadoop-based recommenders randomly down-sample to the interaction histories of power-users. However, this should only have a small impact on the ...
|
|
|
Author: Sebastian Schelter,
2013-03-18, 07:10
|
|
|
Re: Discussion Of ML environment/MR, Mahout - Mahout - [mail # dev]
|
|
...Sean, you were right on this one. I haven't done a thorough benchmark and comparison against GraphLab yet, but I reworked Mahout's ALS code (soon to be committed) to use Multithreaded ...
|
|
|
Author: Sebastian Schelter,
2013-03-16, 07:43
|
|
|
Re: QR decomposition in ALS-WR code - Mahout - [mail # user]
|
|
...Computing an inverse is more costly than solving a linear system via QR decompostion. On 15.03.2013 11:41, Dominik H�bner wrote:...
|
|
|
Author: Sebastian Schelter,
2013-03-15, 10:55
|
|
|
|
