-
Notifications
You must be signed in to change notification settings - Fork 532
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
[FEA] Enable UMAP to build knn graph using NN Descent #5910
[FEA] Enable UMAP to build knn graph using NN Descent #5910
Conversation
Oh the check is failing because we need the updated implementation of RAFT NN Descent (linked above)...! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Great work! Just have a few suggestions. Also, it could be great to add this feature to the simplicial set functions. But, we could do that in another PR too.
Added warning and linked issue for consistency with previous version |
/ok to test |
/ok to test |
/ok to test |
/ok to test |
/merge |
/ok to test |
/ok to test |
/ok to test |
/merge |
Description
build_algo=nn_descent
option to UMAP.Now user can choose the knn graph build algorithm between
"brute_force_knn"
and"nn_descent"
Defaults to
"auto"
, in which case decides to run with brute force knn or nn descent depending on the given dataset size."auto"
decides to run withbrute_force_knn
if either 1) data has <= 50K rows OR 2) data is sparse. Otherwise decides to run withnn_descent
.50K rows roughly chosen based on grid search below. (runtime in ms) - Discussed with Corey
data_on_host
option (defaults to False) when callingfit()
orfit_transform()
Note that brute force knn cannot be used with data on host
Running Benchmarks