CODEWITHSUNDEEP
machine-learningweka
Java_Coder
Java_Coder

Reputation: 55

Weka Machine Learning Package - SVM

I am using the Weka SMO (Support Vector Machines) for Machine Learning. I was trying to optimize the "C" and the "gamma" parameters. I was reading the following documentation:

http://weka.wikispaces.com/Optimizing+parameters

I was trying to follow the "Optimizing SMO with RBFKernel (C and gamma)" section, but I couldn't figure out how I could set the "XProperty" and "YProperty". I would appreciate it if somebody could please let me know how I could optimize the parameters. I was using the GUI tool. Can I use this as part of the weka command for linux.

Thanks.

Upvotes: 2

Views: 1310

Answers (2)

Harsh Patni
Harsh Patni

Reputation: 384

In order to configure these properties you should be selecting "GridSearch" from the classifiers, and proceding with the steps mentioned in the Weka Documentation

You can see the "XProperty" and "YProperty" there.

enter image description here

Also in order to get any weka command you can easily right-click on the Classifier label and copy the configuration files to clipboard.

enter image description here

Upvotes: 1

Bilal Dadanlar
Bilal Dadanlar

Reputation: 820

yes you can use them in command line. they are basically -C and -G parameters in this example:

weka.classifiers.functions.SMO -C 2.0 -L 0.0010 -P 1.0E-12 -N 0 -V -1 -W 1 -K "weka.classifiers.functions.supportVector.RBFKernel -C 250007 -G 0.0"

Upvotes: 2

Related Questions