Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Conversation

glemaitre
Copy link
Member

Follow-up of #19945

Add example of kernel PCA to denoise data as an application.

@ogrisel
Copy link
Member

ogrisel commented Jun 11, 2021

Even if both PCA and kernel PCA have the same MSE, a qualitative analysis will favor the output of the kernel PCA.

I wouldn't be so sure :) It seems that the kPCA reconstructions can be biased into changing 6s into 1s blured with a 6s for instance. I assumed you already tweaked the params a lot and this is the best you could get?

Copy link
Member

@ogrisel ogrisel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok with me although I find the setup a bit artificial as explained in the following:

@ogrisel
Copy link
Member

ogrisel commented Jun 15, 2021

Thanks!

Copy link
Member

@jjerphan jjerphan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a nice example to add.

Here is a quick proofread.

Copy link
Member

@jjerphan jjerphan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thank you @glemaitre !

@ogrisel ogrisel merged commit 1eece2c into scikit-learn:main Jun 29, 2021
samronsin pushed a commit to samronsin/scikit-learn that referenced this pull request Nov 30, 2021

Co-authored-by: Olivier Grisel <[email protected]>
Co-authored-by: Julien Jerphanion <[email protected]>
Co-authored-by: Thomas J. Fan <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants