RT, that's not what I learned... Question posted, see a help... Can help do a better job, completely paid
In single machine commissioning, converting film data set into sequence file:
[hadoop @ localhost hadoop] $hadoop fs - ls - R/mahout1
Rw - r - r - 3 hadoop its 0 2016-04-10 16:09/mahout1 _SUCCESS
Rw - r - r - 3 hadoop its 16:09/mahout1/part 2016-04-10-46935 m - 00000

And then converted to vector file:
[hadoop @ localhost hadoop] $hadoop fs - ls - R/mahout2
DRWXR - xr - x - hadoop its 0 for 2016-04-10/mahout2/df - count
Rw - r - r - 3 hadoop its 0 for 2016-04-10/mahout2/df - count/_SUCCESS
Rw - r - r - 3 hadoop its 2016-04-10 for 313/mahout2/df - count/part - r - 00000
- rw - r - r - 3 hadoop its 225 2016-04-10 16:56/mahout2/dictionary file - 0
Rw - r - r - 3 hadoop its 2016-04-10 for 293/mahout2/frequency. The file - 0
DRWXR - xr - x - hadoop its 0 17:02 mahout2/tf - 2016-04-10 vectors
Rw - r - r - 3 hadoop its 0 17:02 mahout2/tf - 2016-04-10 vectors/_SUCCESS
Rw - r - r - 3 hadoop its 139 17:02 mahout2/tf - 2016-04-10 vectors/part - r - 00000
DRWXR - xr - x - hadoop its 0 2016-04-10 17:05/mahout2/tfidf - vectors
Rw - r - r - 3 hadoop its 0 2016-04-10 17:05/mahout2/tfidf - vectors/_SUCCESS
Rw - r - r - 3 hadoop its 90 2016-04-10 17:05/mahout2/tfidf - vectors/part - r - 00000
DRWXR - xr - x - hadoop its 0 when/mahout2/tokenized - 2016-04-10 documents
Rw - r - r - 3 hadoop its 0 2016-04-10 when/mahout2/tokenized - the documents/_SUCCESS
Rw - r - r - 3 hadoop its 2016-04-10 when 158695/mahout2/tokenized - the documents/part - m - 00000
DRWXR - xr - x - hadoop its 0 2016-04-10 16:56 mahout2/wordcount
Rw - r - r - 3 hadoop its 0 2016-04-10 16:56/mahout2 wordcount/_SUCCESS
Rw - r - r - 3 hadoop its 266 2016-04-10 16:56/mahout2/wordcount/part - r - 00000

Then call fkmeans, output to the mahout4:
[hadoop @ localhost hadoop] $hadoop fs - ls - R/mahout4
DRWXR - xr - x - hadoop its 0 time/mahout4/clusters - 2016-04-10 0
Rw - r - r - 3 hadoop its 207 time/mahout4/clusters - 2016-04-10 0/_policy
- rw - r - r - 3 hadoop its 287 2016-04-10/part time/mahout4/clusters - 0-00000
DRWXR - xr - x - hadoop its 0 yet/mahout4/clusters - 1-2016-04-10 final
Rw - r - r - 3 hadoop its 0 yet/mahout4/clusters - 1-2016-04-10 final/_SUCCESS
Rw - r - r - 3 hadoop its 207 yet/mahout4/clusters - 1-2016-04-10 final/_policy
Rw - r - r - 3 hadoop its 287 yet/mahout4/clusters - 1-2016-04-10 final/part - r - 00000

Then call clusterdump analysis only:
{" identifier ":" SV - 0 "and" r ": []," c ": []," n ":}

Don't know how to do it, if have to call the canopy? The genuflect is begged