Ranking factorization source code

User 875 | 5/4/2015, 11:56:20 AM

Hi Danny,

we need to implement ranking factorization recommender into the spark. Is it possible to ask you because of the source code for the create method for the ranking factorization recommender? How we know, you implement this code in C++, is it possible to have a look into this implementation? Or probably you can recommend us another way to achieve that?


User 91 | 5/4/2015, 5:35:03 PM

Unfortunately, the source code for that particular model isn't part of Dato-Core. All the details about the model are available here: https://dato.com/products/create/docs/generated/graphlab.recommender.rankingfactorizationrecommender.RankingFactorizationRecommender.html#graphlab.recommender.rankingfactorizationrecommender.RankingFactorizationRecommender

We have put a lot of effort into making the model training robust so it might take a bit of work into porting it over to Spark. Is there any reason why you want to implement it in Spark and not use GraphLab create?

User 875 | 5/6/2015, 9:05:24 AM

because the Spark can manipulate the bigger amount of the data