You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Found an error in HDBScan code when add auto-weighting function:
When compute the core-distances for each observation, the euclidean distance is used. However, the squared euclidean distance is used when search neighbors. This is wrong, and makes the parameter "Min samples", which defines the core-distance always "smaller".
The text was updated successfully, but these errors were encountered:
Found an error in HDBScan code when add auto-weighting function:
When compute the core-distances for each observation, the euclidean distance is used. However, the squared euclidean distance is used when search neighbors. This is wrong, and makes the parameter "Min samples", which defines the core-distance always "smaller".
The text was updated successfully, but these errors were encountered: