-
Notifications
You must be signed in to change notification settings - Fork 547
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
[REVIEW] Provide access to cuml.DBSCAN core samples #2499
[REVIEW] Provide access to cuml.DBSCAN core samples #2499
Conversation
…AN core samples
…AN core samples
… into fea-dbscan-core-samples
Please update the changelog in order to start CI tests. View the gpuCI docs here. |
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.
Overall LGTM. Just one minor comment.
Also tagging @MatthiasKohl in case he wants to look at this.
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.
Changes LGTM.
rerun tests |
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.
Changes on the Python side lgtm
Closes #2311
Would like feedback on Python DBSCAN class constructor argument
calc_core_sample_indices
that I added. This option allows users who won't ever usecore_sample_indices_
to avoid any additional computation cost. Defaults to True to match scikit-learn.Also, the C API functions
cumlSpDbscanFit
andcumlDpDbscanFit
incpp/src/dbscan/dbscan_api.h
signatures were changed to account for the newcore_sample_indices
array without needing 2 additional functions with slightly different names. Users can easily passNULL
forcore_sample_indices
but it would be a breaking change.