scikit learn - python sklearn KDTree with haversine distance -

i try create kd tree of wgs84 coordinates , find neighbors within radius

from sklearn.neighbors.dist_metrics import distancemetric sklearn.neighbors.kd_tree import kdtree     t = kdtree([[47.8665, 8.90123]], metric=distancemetric.get_metric('haversine')) 

but following error:

valueerror: metric haversinedistance not valid kdtree 

how can use haversine distance in kd-tree?

the k-d-tree can (to best of knowledge) used minkowski norms.

there other trees such ball tree in sklearn, or covertree in elki work haversine distance because metric.