Scorer node doesn't work

I am new to KNIME and am trying to build a simple Naive Bayes model using a dataset having 48842 instances. I am trying to execute the model on Windows 10 (8GB of RAM total in my laptop).

Model is like this -  File Reader->Partitioning->NB Learner->NB Predictor->Scorer

All the nodes, except the scorer node works. On executing the Scorrer node it proceeds to 50% and then hangs. I am not sure what the problem migh be.

Have made some changes to the knime.ini file based on some of the forum suggestions...


Also, selected "write tables to disk" for the scorer config options. However, with no success...

Earlier I was getting the Java Heap error/ GC there are no errors displayed on the console...however the application still freezes!!


Can anybody help please? This is the first model I am trying to build and without success so far.









How many different classes do you have in your dataset? And did you select the correct columns in the scorer's dialog? If you have an extremly large number of classes or have selected the wrong column the node tries to compute statistics for all of them which may require lots of memory.

Hi Thor,

It is the "Census Income" dataset available from UCI.....

there are only 2 classes with 14 attribiutes...and yes, I have also checked the columns in the scorer's dialog but will recheck as you suggest...

Please let me know if there is anything else that I should check on?





We are using this dataset quite often and it works without any problems. The only reason I can think of that you selected one column with many different values in the dialog (>>1000). Then a huge confusion matrix is computed which will eat up all the memory. Maybe you can quickly share your configuration.

You are so correct...It was because I had selected a wrong column in the scorer....Now the model works perfectly...


Thank you so much!!