Familiarity with graphlab.factorization_recommender.create

User 3274 | 4/29/2016, 1:02:42 PM


I was using the above mentioned module on a dataset for prediction of sales of items across various outlets. I have dealt this problem by incorporating a matrix factorization style solution. However I am not sure as to how I would be able to use the cross validation toolkit for this purpose. Any guidance on this would be really helpful. Thank you, Prithviraj Dutta


User 16 | 4/30/2016, 2:16:56 AM

@prithviraj - here are a couple of links you may find useful.

We have an example in our user guide about using cross validation to tune a model: https://dato.com/learn/userguide/modelparametersearch/models.html

There are also the API docs: https://dato.com/products/create/docs/generated/graphlab.toolkits.crossvalidation.crossvalscore.html#graphlab.toolkits.crossvalidation.crossvalscore

Hopefully that helps. Let me know if you have any specific questions.

Thanks, Toby