Parallelize init_bound_dense in Elkan algorithm #19052
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks @YusukeNagasaka. looks good. It would make sense to also do the change in |
ogrisel
approved these changes
Dec 21, 2020
|
This makes huge performance improvement on A64FX CPU, even not much improvement on Intel CPU. @YusukeNagasaka will change |
|
The same parallelization is applied to The schedule in prange I adopted is |
|
Merged! Thank you very much @YusukeNagasaka! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this implement/fix?
Parallelize init_bound_dense function. This is an initialization part in Elkan algorithm.
Any other comments?
This fix does not affect the quality of the clustering itself. This parallelization only reduces execution time of KMeans.fit on multi-core CPUs, even more beneficial on many-core CPUs.
For instance, I tested on a machine with 2 sockets of Intel Xeon CPUs (totally 40 cores). The table below is the time spent during KMeans.fit and init_bound_dense (the time in seconds). The data is generated uniformly at random with
n_samples=1M, n_features=100. The parameters of KMeans aren_clusters=1000, init='random', algorithm='elkan', max_iter=1000.