mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Owen <sro...@gmail.com>
Subject Re: R: The best evaluator for recommendations in binary data sets
Date Wed, 29 Jul 2009 08:23:19 GMT
Yes, all this is done for you. The training set contains everything
except a few items for one user (see they are removed on about line
179? my code is a bit different).

On Wed, Jul 29, 2009 at 8:35 AM, Claudia Grieco<grieco@crmpa.unisa.it> wrote:
>> For example maybe you take away three of the
> user's items and see if the recommender recommends 3 of them back
>
> I have looked at the code...does the GenericRecommenderIRStatsEvaluator class do this
stuff itself?
> I see that it creates a training set, but it appears from the code that this set contains
all items...

Mime
View raw message