Open Maryom opened 6 years ago
Hi,
Could you please describe how a bucket kd-tree is supposed to work?
Thanks!
Thanks for your fast reply.
Kindly, check this document:
Bucket means points are collected in leaves until they overflow. Only then are they split. This makes the tree more likely to be balanced since we can choose a smarter splitting point than whatever we receive first.
Do you think it will be easy to add bucket to this repo. ?
Thanks.
The fact that each node corresponds to one point (and has up to two sub nodes) is a basic assumption of the kdtree
module. Such a change would probably affect many parts of the whole data structure.
As I do not have sufficient time at the moment, I will not be able to implement this in the foreseeable future, I would however be available for reviewing any pull requests.
Ok thanks. Maybe I will pull a request to do it. I'm thinking and I'm searching.
Hi,
Thanks for your repo.
I'm wondering it is possible to add bucket kd-trees ? I found Matlab code of it. However, I need it in python.
Thanks.