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

DBSCAN warning about batch size #4449

Open AmenRa opened 2 years ago

AmenRa commented 2 years ago

Hi, BDSCAN prints this warning, but I do not find any information on fixing what's wrong.

Batch size limited by the chosen integer type (4 bytes).
Using the larger integer type might result in better performance

How can I change the integer type?

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.

amritphy commented 2 years ago

I am facing the same warning. Can anyone please comment on this? @robertmaynard

robertmaynard commented 2 years ago

I am facing the same warning. Can anyone please comment on this? @robertmaynard

I am not famillar with dbscan, so I can't tell you if there is anything you can do to remidy the situation.

achirkin commented 2 years ago

According to the documentation, I guess you can try to supply 'int64' as the second argument of your fit call.

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.