Got a YouTube account?

New: enable viewer-created translations and captions on your YouTube channel!

English subtitles

← Match Points with Clusters - Intro to Machine Learning

Get Embed Code
4 Languages

Showing Revision 3 created 05/25/2016 by Udacity Robot.

  1. In k-means, you randomly draw cluster centers and
  2. say our first initial guess is, say, over here and over here.
  3. These are obviously not the correct cluster centers.
  4. You're not done yet.
  5. But k-means now operates in two steps.
  6. Step number is assign and step number two is optimize.
  7. So let's talk about the assignment.
  8. For classes in number one, I want you to click on exactly those of
  9. the red points that you believe are closer to center one than center two.