How to interpret SVM Learner output

Hi,

I got the following output from a SVM Learner.

I tried but couldn’t find an explanation for the Lower and Upper Bound output.

Also, is there anywhere I can see the weights of the attributes in the SVM model?

Thanks!

Hi,
In this view you only see the spec of the training data. More interesting is the PMML document in the first tab. It contains the actual model that was learned. Please see here for a description of the model.
Kind regards
Alexander

3 Likes

Hi Alexander,

Thanks for the reply.

I’ve attached a screenshot of the PMML document.

Am I right to assume (as shown in the arrow in the image) that Support Vector Rock_Row0 is associated with the first coefficient? Is there a way to combine these two information together in a tabular format for easy viewing and usage?

Thanks in advance!

Hi,
You are correct, each vector has one coefficient and they are associated as you guessed. To extract the information into a table, you can use the PMML to Cell node followed by an XPath node.
Kind regards
Alexander

4 Likes

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.