Cannot agree more with your words. Could you add one section about "how and what to contribute" to MLlib's guide? -Xiangrui
On Mon, Apr 21, 2014 at 1:41 PM, Nick Pentreath <nick.pentre...@gmail.com> wrote: > I'd say a section in the "how to contribute" page would be a good place to > put this. > > In general I'd say that the criteria for inclusion of an algorithm is it > should be high quality, widely known, used and accepted (citations and > concrete use cases as examples of this), scalable and parallelizable, well > documented and with reasonable expectation of dev support > > Sent from my iPhone > >> On 21 Apr 2014, at 19:59, Sandy Ryza <sandy.r...@cloudera.com> wrote: >> >> If it's not done already, would it make sense to codify this philosophy >> somewhere? I imagine this won't be the first time this discussion comes >> up, and it would be nice to have a doc to point to. I'd be happy to take a >> stab at this. >> >> >>> On Mon, Apr 21, 2014 at 10:54 AM, Xiangrui Meng <men...@gmail.com> wrote: >>> >>> +1 on Sean's comment. MLlib covers the basic algorithms but we >>> definitely need to spend more time on how to make the design scalable. >>> For example, think about current "ProblemWithAlgorithm" naming scheme. >>> That being said, new algorithms are welcomed. I wish they are >>> well-established and well-understood by users. They shouldn't be >>> research algorithms tuned to work well with a particular dataset but >>> not tested widely. You see the change log from Mahout: >>> >>> === >>> The following algorithms that were marked deprecated in 0.8 have been >>> removed in 0.9: >>> >>> From Clustering: >>> Switched LDA implementation from using Gibbs Sampling to Collapsed >>> Variational Bayes (CVB) >>> Meanshift >>> MinHash - removed due to poor performance, lack of support and lack of >>> usage >>> >>> From Classification (both are sequential implementations) >>> Winnow - lack of actual usage and support >>> Perceptron - lack of actual usage and support >>> >>> Collaborative Filtering >>> SlopeOne implementations in >>> org.apache.mahout.cf.taste.hadoop.slopeone and >>> org.apache.mahout.cf.taste.impl.recommender.slopeone >>> Distributed pseudo recommender in >>> org.apache.mahout.cf.taste.hadoop.pseudo >>> TreeClusteringRecommender in >>> org.apache.mahout.cf.taste.impl.recommender >>> >>> Mahout Math >>> Hadoop entropy stuff in org.apache.mahout.math.stats.entropy >>> === >>> >>> In MLlib, we should include the algorithms users know how to use and >>> we can provide support rather than letting algorithms come and go. >>> >>> My $0.02, >>> Xiangrui >>> >>>> On Mon, Apr 21, 2014 at 10:23 AM, Sean Owen <so...@cloudera.com> wrote: >>>>> On Mon, Apr 21, 2014 at 6:03 PM, Paul Brown <p...@mult.ifario.us> wrote: >>>>> - MLlib as Mahout.next would be a unfortunate. There are some gems in >>>>> Mahout, but there are also lots of rocks. Setting a minimal bar of >>>>> working, correctly implemented, and documented requires a surprising >>> amount >>>>> of work. >>>> >>>> As someone with first-hand knowledge, this is correct. To Sang's >>>> question, I can't see value in 'porting' Mahout since it is based on a >>>> quite different paradigm. About the only part that translates is the >>>> algorithm concept itself. >>>> >>>> This is also the cautionary tale. The contents of the project have >>>> ended up being a number of "drive-by" contributions of implementations >>>> that, while individually perhaps brilliant (perhaps), didn't >>>> necessarily match any other implementation in structure, input/output, >>>> libraries used. The implementations were often a touch academic. The >>>> result was hard to document, maintain, evolve or use. >>>> >>>> Far more of the structure of the MLlib implementations are consistent >>>> by virtue of being built around Spark core already. That's great. >>>> >>>> One can't wait to completely build the foundation before building any >>>> implementations. To me, the existing implementations are almost >>>> exactly the basics I would choose. They cover the bases and will >>>> exercise the abstractions and structure. So that's also great IMHO. >>>