mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lance Norskog <goks...@gmail.com>
Subject Re: Help using mahout for k-means clustering on existing vectors
Date Thu, 12 Jan 2012 01:24:22 GMT
Does an empty entry mean a zero value? If so, try changing the text
file to use 0 instead of the empty space.

Is there a text line which has 16 numbers and the rest empty spaces?
And another text line with 91 numbers?

On Wed, Jan 11, 2012 at 2:31 AM, rahul raghavendhra
<rahulraghavendhra1@gmail.com> wrote:
> Hi all,
>
>  i have run that org.apache.mahout.clustering.syntheticcontrol.<>.Job
> successfully..
>
>  when i run with similar dataset(double values separated by ' ' (space))..
>
> i got the error  org.apache.mahout.math.CardinalityException: Required
> cardinality 16 but got 91
>
> How this Cardinality is calculated and how it is passed  to kmeans driver..
> how to calculate the cardinality for any dataset ?
>
> please help
>
>
>
>
> ./rahul
>
>
> On Tue, Jan 10, 2012 at 9:31 AM, Grant Ingersoll <gsingers@apache.org>wrote:
>
>> The CSVVectorIterator will get you vectors from a CSV file, then you just
>> need to write them out to the SequenceFile.  All you need is a driver that
>> wraps the SequenceFileVectorWriter and calls the write method.
>>
>>
>> On Jan 9, 2012, at 2:50 PM, Daniel Quach wrote:
>>
>> > I have a file of vectors I formulated in csv format, and I want to use
>> mahout to perform k-means clustering on the vectors in this file.
>> >
>> > However, it seems mahout expects the input data to be formatted in a
>> SequenceFile format, and I'm not sure if there's a way to easily do this
>> (are there existing tools?)
>>
>> --------------------------------------------
>> Grant Ingersoll
>> http://www.lucidimagination.com
>>
>>
>>
>>



-- 
Lance Norskog
goksron@gmail.com

Mime
View raw message