cuml icon indicating copy to clipboard operation
cuml copied to clipboard

[FEA] Soft clustering with HDBSCAN

Open jhfoxliu opened this issue 4 years ago • 4 comments

Is your feature request related to a problem? Please describe. I wish using HDBSCAN soft clustering method (which is implemented in the CPU version, see https://hdbscan.readthedocs.io/en/latest/soft_clustering.html) with cuML.

Describe the solution you'd like It is the best to update a "all_points_membership_vectors" function the same as its CPU version.

Describe alternatives you've considered It's can be OK if the attributes (prediction_data, raw_tree, etc.) necessary for soft clustering are provided. With that attributes, I can use CPU version to compute the values. However, they are missing in cuML.

Additional context

jhfoxliu avatar Dec 31 '21 17:12 jhfoxliu

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.

github-actions[bot] avatar Jan 30 '22 18:01 github-actions[bot]

I would be interested in this as well.

markselias avatar Mar 31 '22 12:03 markselias

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.

github-actions[bot] avatar Apr 30 '22 14:04 github-actions[bot]

This issue has been labeled inactive-90d due to no recent activity in the past 90 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed.

github-actions[bot] avatar Jul 29 '22 15:07 github-actions[bot]

With the merge of #4800, soft clustering the original dataset with all_points_membership_vectors is now available. Please give it a try and file issues if you run into any issues or have any feedback.

Closing this issue as resolved. Soft-clustering a new set of points with membership_vector is being tracked in https://github.com/rapidsai/cuml/issues/4724

beckernick avatar Aug 29 '22 13:08 beckernick