0
$\begingroup$

I am trying to learn to use libsvm, I have used some defalut training sets to generate the plots by grid-search and obtained a plot below.

I want to know,

  • What type of plot is this,

  • What does it really tells us,

  • How to understand and analyze this plot.

    Thanks in advance.

Libsvm grid-search

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

Grid search performs a search in two parameters of your svm: c and gamma.

These two parameters are displayed in logarithmic axes x and y in this graph. The lines indicate which areas the grid search has been.

The colours of the lines are like "hight-lines" on a geographical map. They indicate the quality of the solutions found there and display the graphical bounds of the searched space in a graph.

Remember how grid search works:

The initial grid is worked on with 2-fold CV to determine the values of the parameter pairs for the selected type of evaluation (e.g., accuracy). The best point in the grid is then taken and a 10-fold CV is performed with the adjacent parameter pairs. If a better pair is found, then this will act as new center and another 10-fold CV will be performed (kind of hill-climbing). This process is repeated until no better pair is found or the best pair is on the border of the grid. In case the best pair is on the border, one can let GridSearch automatically extend the grid and continue the search. -- source

If I am correct, the lines indicate where the deeper search is performed.

But just to double check, which software did you use to generate this? Matlab or weka?

Improve this answer
$\endgroup$
1
  • $\begingroup$ I was using grid.py from libsvm package $\endgroup$
    – aravind ramesh
    Commented May 10, 2014 at 6:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.