Rapid Miner Go Error: Scoring Data Attributes must be the subset of the original attributes!

cramsdencramsden Member Posts: 42 Contributor I
edited August 2020 in Help
Hi, I've been using rapid miner go for a while now and am getting an error that I have never seen before.   I am trying to use a model ive built to get make predictions on new data. I have done this a number for times with no problems before but today I am getting this error

'undefined Error: File AugU0to5.xlsx has invalid content. Scoring Data Attributes must be the subset of the original attributes!'

every time i try to upload new data.  I am even trying to run a prediction on the exact same spreadsheet I used on Friday with no problems but today I am getting this error.

How do I fix this?
Tagged:

Best Answer

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist
    Hi,

    usually this means, that the excel file you want to score has additional columns which weren't there during training? Did you check this?

    Best,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • cramsdencramsden Member Posts: 42 Contributor I
    Yes, usually it just ignores any additional columns though.   I have also tried scoring a spreadsheet on a model I made last week and scored without issue, (I made no changes to it) and I get this error now.
  • cramsdencramsden Member Posts: 42 Contributor I
    I'm trying to score a sheet now that has absolutely identical columns as training and still getting this error
  • cramsdencramsden Member Posts: 42 Contributor I
    On other thing to note, the only data I am able to upload to score is the exact same sheet with all the data I used to train the model, which is kind of pointless.  It is rejecting any new data sets.   It looks like something might have changed over the weekend where its not allowing any different data points to be scored.   
  • cramsdencramsden Member Posts: 42 Contributor I
    The only way I can get it to work now is over writing the spreadsheet used to train the model with new data and then uploading that.  Any fresh spreadsheet even if the columns match I get this error. That wasnt the case last week
  • cramsdencramsden Member Posts: 42 Contributor I
    Yup, seems to work like before now!  Thanks guys!
Sign In or Register to comment.