Random forest: Target column does not have values assigned.

KNIME Analytics Platform
1

Hi folks,

On the Random Forest learner:

I have data with 15 value columns and 1 value column that needs to be predicted. The attributes have been read as Double, the target as String (to define a nominal value). However, the Random Forest learner, does not accept the data, shwoing the following error message: Execute failed: The target column does not have possible values assigned. Most likely it has too many different distinct values (learning an ID column?) Fix it by preprocessing the table using a "Domain Calculator".

It does not matter what data is put in: neither the original values [0,1] nor the autoscaled values (so the ensemble avg=0, std=1) is accepted. I would appreciate to get some clues.

Please do provide more helpful clues: the error message is not helpful at all, because it's absolutely a miracle why a value with a continuous range of double value is considered invalid input.

 

Thanks, 

Ruud

2

If you read the error message carefully it tells you that the problem is the target column and not any of the attribute columns. Did you insert a Domain Calculator node before the Random Forrest Learner as suggested by the message?