Using the Mahout CLI is not a typical approach to research features though it might be useful.
What I always ask is: what is the feature, what does it indicate about the user, how many possible items for the feature type, how many users have indicated this feature. average users/item and average items/user for the feature are also useful when compared to the same for conversion or primary feature/indicator. On May 19, 2017, at 9:20 AM, Pat Ferrel <p...@occamsmachete.com> wrote: Some ideas are left on the SO question On May 18, 2017, at 6:14 PM, Daniel Gabrieli <dgabri...@salesforce.com <mailto:dgabri...@salesforce.com>> wrote: Hi, I am trying to research potential features to feed into the Universal Recommendation engine by using the mahout command line interface. I can successfully run mahout spark-rowsimilarity <https://mahout.apache.org/users/recommender/intro-cooccurrence-spark.html> on up to about 500 rows of data. After that I get the error below: INFO DAGScheduler: Job 7 failed: saveAsTextFile at TextDelimitedReaderWriter.scala:294, took 1.110184 s Exception in thread "main" org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 12.0 failed 1 times, most recent failure: Lost task 0.0 in stage 12.0 (TID 24, localhost): java.lang.IllegalArgumentException I tried various cli options like below but get the same error: mahout spark-rowsimilarity --maxObservations 500000 -sem 6g -ma "local[4]" --input items.csv --output /tmp/output Any suggestions would be most helpful. I cross posted on Stack Overflow here <http://stackoverflow.com/questions/44059463/mahout-spark-rowsimilarity-error> Thanks -- You received this message because you are subscribed to the Google Groups "actionml-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to actionml-user+unsubscr...@googlegroups.com <mailto:actionml-user+unsubscr...@googlegroups.com>. To post to this group, send email to actionml-u...@googlegroups.com <mailto:actionml-u...@googlegroups.com>. To view this discussion on the web visit https://groups.google.com/d/msgid/actionml-user/1C2E66DF-BD3B-4584-B30B-5934D0E90B84%40occamsmachete.com <https://groups.google.com/d/msgid/actionml-user/1C2E66DF-BD3B-4584-B30B-5934D0E90B84%40occamsmachete.com?utm_medium=email&utm_source=footer>. For more options, visit https://groups.google.com/d/optout <https://groups.google.com/d/optout>.