8000 New feature: NCA · Issue #3213 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content

New feature: NCA #3213

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
RolT opened this issue May 28, 2014 · 5 comments
Closed

New feature: NCA #3213

RolT opened this issue May 28, 2014 · 5 comments

Comments

@RolT
Copy link
RolT commented May 28, 2014

I would like to contribute by adding an algorithm performing the Neighborhood component analysis (http://www.cs.toronto.edu/~fritz/absps/nca.pdf)

Basically, it's trying to learn a distance metric by optimizing the leave-one-out performance on the training data. In a first time, I would first only follow the paper, and then try to add different solving methods in the parameters.

Tell me if you would be interested, if have any recommendations for the architecture or if someone else is already doing this.

@agramfort
Copy link
Member

metric learning methods would be neat to have, although I am not sure we
should start with NCA. Thoughts anyone?

@RolT
Copy link
Author
RolT commented Jun 2, 2014

Ok, I started implementing it in python. I will try to follow the structures used in scikit-learn so that it will be easy to add if some people are interested.

@sreenivasraghavan71
Copy link

I am Interested in NCA implementation. Can you tell me till what stage has work been completed and what additional algorithms have to be implemented?

@amueller
Copy link
Member

Please have a look at #4789.

@RolT
Copy link
Author
RolT commented Jul 31, 2015

I stopped working on that feature. I did an python implementation following the scikit-learn API, but I did not intergrate it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants
0