bug in Gaussian Distributed Assigner?

KNIME Analytics Platform
#1

I'm using the Gaussian Distributed Assigner, using a categorical dependency column as per the documentation for three categories. I am finding the output is shuffled between the categories.

I have three categories, target, high and low. The mean and standard deviations for the three are 

target: 0.12 mean, 0.07 std dev

high: 0.17 mean, 0.09 std dev

low: 0.09 mean, 0.035 std dev

 

the output for 10,000 numbers per category is

target: 0.09, 0.035

high: 0.119, 0.07

low: 0.171, 0.09

 

Clearly the generation is according to the distribution mean and standard deviation, but the assignment to the category is wrong.

Is this a bug or is there something specific I need to do to get the proper mapping to the category?

I'm attaching the workflow.

David

 

 

#2

Hi David,

thank you for reporting this. We already fixed the bug with 3.4.2.

I played with your workflow, and I needed to enter a new Gaussian Dist. node and reconfigure it, but than it was working correctly. 

Which KNIME Version are you currently using?

Best, Iris 

#3

thanks, I'm still on 3.2.2 on my desktop. Looking forward to upgrading it.

Best

David