The RapidMiner community is on read-only mode until further notice. Technical support via cases will continue to work as is. For any urgent licensing related requests from Students/Faculty members, please use the Altair academic forum here.

speeding up Numerical To Binomial?

lansuminclansuminc Member Posts: 9 Contributor II
edited November 2019 in Help
After some pre-processing I ended up with a big table of 0's and 1's. These have type real. But I want them to become binomials.
The operator Numerical To Binomial does the trick... slowly. For my table of (test!) data (1.5mm rows of 6k columns) it is now taking 2.5 hours and no sign of it being complete.

Does anyone have a trick to speed things up?

One thing I thought of is writing the table to a temp CSV file and then use the smart guessing of the CSV import to make them binomial.
I've tried the GuesTypes operator, but that I killed after having run for an hour or so...
Writing to a tmp database table does not work as my postgresql database does not like more than 1600 columns.
Tagged:

Answers

  • awchisholmawchisholm RapidMiner Certified Expert, Member Posts: 458 Unicorn
    Try the Materialize Data operator immediately before the slow operator.

    Regards

    Andrew
Sign In or Register to comment.