Code covered by the BSD License  

Highlights from
Mean shift for finding modes

Be the first to rate this file! 21 Downloads (last 30 days) File Size: 4.66 KB File ID: #39079
image thumbnail

Mean shift for finding modes

by

 

19 Nov 2012 (Updated )

The submission finds modes in data. Data is generated from a mixture of gaussian with added outliers

| Watch this File

File Information
Description

Run plotModes() to see result.
Note: 40% generated data is added outlier.

Correction:
In findModes.m, go to kernelized_weights() function and replace:
euclidean_dist = sum(offset.*offset,2) .* inv_bandwidth_sq;

with:
euclidean_dist = sum(offset.*offset,2);

MATLAB release MATLAB 8.0 (R2012b)
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (1)
19 Nov 2012 Soumitry J Ray

Correction:
In findModes.m, go to kernelized_weights() function and replace:
euclidean_dist = sum(offset.*offset,2) .* inv_bandwidth_sq;

with:
euclidean_dist = sum(offset.*offset,2);

Updates
27 Nov 2012

Description change

Contact us