How to fine-tune the sensitivity of Bagging parameters using Weka extensions?

[pippo]

Sorry, but I’ve just started using KNIME, so I’m not very familiar with it yet. As mentioned in the title, I would like to understand how to fine-tune sensitivity using the Weka extensions. Normally, I would use a Parameter Optimization Loop Start, X-Aggregator, X-Partitioner, and Parameter Optimization Loop End. How can I apply this to fine-tune the parameters of the Bagging node?

[ActionAndi]

Hi Pippo,

have you tried to adapt the solution I provided you? I think you must only adjust ne parameter naming, range and if needed add some more parameters.

241227_OptimML_by_Sens – actionandi

Andi

[pippo]

Hi thank you for the reply, but I wanted to use the bagging classifier of weka and I did not find any resources(workflow examples, doc…). This is why I asked an another question

Bagging (3.7)

[ActionAndi]

Yeah, there aren’t (m)any example workflows, but it’s quite straight forward:

In the Bagging-Node you can adjust Model-Types, Number of Models etc.

[pippo]

Sorry but I did not write the title and explain correctly but I wanted to know how to fine tune the sensitivity using cross validation of the bagging model

[ActionAndi]

I think approach is very the same like in my example above:

Heres the updated workflow:

241229_OptimML_by_Sens – actionandi