mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Owen <sro...@gmail.com>
Subject Re: running examples for mahout.
Date Sat, 25 Sep 2010 07:54:02 GMT
The output of "GroupLensRecommenderEvaluatorRunner"? That is the
result of an evaluation of how good the predicted ratings are, for a
given algorithm and given data set. You can change the algorithm it is
evaluating and see a different number. I believe it is using
average-absolute-difference evaluation. So the final number is just
how different the rating was from predicted, on average.

No, the result of this tool is not recommendations. But, input would
also be users and items and ratings if you wanted to do
recommendations. Why is that surprising? the whole idea of
recommenders is to generate additional user/item/rating tuples that
don't already exist.

On Sat, Sep 25, 2010 at 7:01 AM, web service <wbsrvc@gmail.com> wrote:
> Ok. Got it running. But what is the output expected to be. I suppose I can
> find some recommendations ? But the dataset is already of the users and
> corresponding ratings ? A bit confused about it.
>
> Exactly what does output mean or what does it look like ?
>
>
> -Mac
>
> On Fri, Sep 24, 2010 at 12:25 PM, Sean Owen <srowen@gmail.com> wrote:
>
>> The GroupLensDataModel expects the input from the "1 million" data set
>> which has a different format. Use that instead. Or to make this work,
>> just modify the code this runs here to not use GroupLensDataModel, but
>> instead a FileDataModel, which will parse this more normal input.
>>
>> I guess this bit of info was not brought over to this wiki. I'll have
>> to update it.
>>
>> On Fri, Sep 24, 2010 at 6:15 PM, web service <wbsrvc@gmail.com> wrote:
>> > The input file can be obtained at -
>> > http://www.grouplens.org/system/files/ml-data.tar__0.gz
>> >
>> > A sample entry in u.data looks like this
>> > [code]
>> > 196 242 3 881250949
>> > [/code]
>> >
>> > I just want to run example. That's it.
>> >
>> > -Mac
>> >
>> > On Fri, Sep 24, 2010 at 9:28 AM, Sean Owen <srowen@gmail.com> wrote:
>> >
>> >> Looks like the input file's format isn't right. It's expecting that
>> >> funky "::"-delimited file. What is the input like?
>> >>
>> >> The error message sure could be better, I can fix that.
>> >>
>> >> On Fri, Sep 24, 2010 at 4:54 PM, web service <wbsrvc@gmail.com> wrote:
>> >> > How do I run examples listed at
>> >> > https://cwiki.apache.org/MAHOUT/recommendationexamples.html
>> >> >
>> >> > I downloaded the file for grouplens dataset and after running the
>> >> command, I
>> >> > get back an error -
>> >> > [code]
>> >> >
>> >> > mvn -e exec:java
>> >> >
>> >>
>> -Dexec.mainClass="org.apache.mahout.cf.taste.example.grouplens.GroupLensRecommenderEvaluatorRunner"
>> >> > -Dexec.args="-i
>> >> > ~/software/hadoop/mahout/mahout-0.3/examples/input/ml-data/u.data"
>> >> >
>> >> >
>> >> >
>> >> > .................
>> >> > Caused by: java.lang.StringIndexOutOfBoundsException: String index
out
>> of
>> >> > range: -1
>> >> >    at java.lang.String.substring(String.java:1937)
>> >> >    at
>> >> >
>> >>
>> org.apache.mahout.cf.taste.example.grouplens.GroupLensDataModel.convertGLFile(GroupLensDataModel.java:61)
>> >> >    at
>> >> >
>> >>
>> org.apache.mahout.cf.taste.example.grouplens.GroupLensDataModel.<init>(GroupLensDataModel.java:48)
>> >> >    at
>> >> >
>> >>
>> org.apache.mahout.cf.taste.example.grouplens.GroupLensRecommenderEvaluatorRunner.main(GroupLensRecommenderEvaluatorRunner.java:49)
>> >> >
>> >> > [/code]
>> >> >
>> >> >
>> >> > -Mac
>> >> >
>> >>
>> >
>>
>

Mime
View raw message