Weka predictor crashes with NayvBayse Multinomial node

KNIME Analytics Platform
#1

Hi,
I am using the weka predictor node with different classifier.
It works well with all of them except the NayvBayse Multinomial node: I got the following error:
ERROR Weka Predictor (3.7) 2:541
ERROR Weka Predictor (3.7) 2:541 Execute failed: java.lang.NullPointerException

Is there a reason for this error and how can I get rid of it to get the predicted values?
Thanks,
Zied

#2

I was not able to reproduce this issue. Can you share your workflow with us? And can you additionaly provide the respective log?

#3

I think I found the reason of the bug: it is because of the presence of missing values.
Thanks for your help.
Zied

1 Like
#4

Thanks for following up on this! Indeed, the Weka Naive Bayes Multinomial classifier can not handle missing data. We will make sure to provide a more descriptive error message for this case.

#5