storpipfugl / pykdtree

Fast kd-tree implementation in Python
GNU Lesser General Public License v3.0
215 stars 47 forks source link

Missing query_ball_point #23

Closed akaszynski closed 6 years ago

akaszynski commented 7 years ago

Hello,

This is an excellent algorithm that's much faster than scipy for my nearest neighbor searches. Thanks!

Would it be too much work to add query_ball_point that's currently in the scipy implementation?

Alex

akaszynski commented 6 years ago

Turns out that you can do much of the same functionality by setting the distance_upper_bound= to some number and then ignoring neighbors with infinite distance. You still have to set the number of neighbors to find, but there's not too much of a time penalty to that.

Mechazo11 commented 2 years ago

Turns out that you can do much of the same functionality by setting the distance_upper_bound= to some number and then ignoring neighbors with infinite distance. You still have to set the number of neighbors to find, but there's not too much of a time penalty to that.

@akaszynski can you share the code that you used to replicate the functionality of query_ball_point() method?