Skip to content
This repository has been archived by the owner on Nov 19, 2020. It is now read-only.

KNearestNeighbors parallel #401

Closed
datashinobi opened this issue Jan 30, 2017 · 2 comments
Closed

KNearestNeighbors parallel #401

datashinobi opened this issue Jan 30, 2017 · 2 comments

Comments

@datashinobi
Copy link

Is KNearestNeighbors parallel?

Can it take a custom distance function and run in multi threads
@cesarsouza
Copy link
Member

Hello there,

The current implementation of the k-Nearest Neighbors can be found here. Unfortunately, as it can be seen, it does not use multiple threads (yet).

However, it should be fairly straightforward to replace the standard loops in the algorithm by Parallel.For calls. I can update it in some days. But since it might be simple to implement it, please also feel free to send a pull request or ask more information on how to do it!

Regards,
Cesar

cesarsouza added a commit that referenced this issue Feb 2, 2017
@cesarsouza
Updating K-Nearest Neighbor to use the new classification/learning in…
6c2dfe1 
…terfaces;

Adding parallelism to the non-data-structure-based version of the algorithm.

Updates GH-401: KNearestNeighbors parallel.
@cesarsouza
Copy link
Member

Fixed in 3.5.0.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@cesarsouza @datashinobi