This software contains several matlab scripts for computing the RDE (relevant dimensionality estimate). The RDE measures the number of leading PCA components in feature space which contain the relevant information about the labels in a supervised learning problem.
The package also contains methods for plotting the kernel PCA components, performing model selection based on the RDE, and visualizing the RDE for different kernel parameters. Implementations of linear, polynomial, Gaussian, and rational quadratic kernels are also contained.
For more information, have a look at the README file and at the rdetools.m script, which contains a small demo of how to use the routines.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.