search

rapidsai / cuml

cuML - RAPIDS Machine Learning Library
https://docs.rapids.ai/api/cuml/stable/
Apache License 2.0
4.2k stars 525 forks source link

[FEA] Would DBSCAN supports being performed on separate datasets in a single call? #4600

Open georgeliu95 opened 2 years ago

georgeliu95 commented 2 years ago

Is your feature request related to a problem? Please describe. I wish I could use cuML to do DBSCAN clustering on separate datasets (with same shapes) in a single call.

Describe the solution you'd like For example, we pass a dense feature with shape (n_batches, n_samples, n_features) to DBSCAN().fit() once, instead of make n_batches calls where each call fits (n_samples, n_features) feature.

Describe alternatives you've considered Make several calls with different CUDA streams, and ensure it would not be blocked.

github-actions[bot] commented 2 years ago

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.

divyegala commented 2 years ago

@georgeliu95 I'm tagging this as a feature request. This is not currently possible, but we will explore it in the upcoming releases

github-actions[bot] commented 2 years ago

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] commented 2 years ago

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.