Hi everyone,
i need to perform adaptive binning of multidimensional data, in an online fashion.
as far as i understand i need a dynamic kd-tree,
is there something in the julia ecosystem that implements a structure resembling what i need?
Hi everyone,
i need to perform adaptive binning of multidimensional data, in an online fashion.
as far as i understand i need a dynamic kd-tree,
is there something in the julia ecosystem that implements a structure resembling what i need?
Nothing that I’m aware of, but you might find something that suits your needs in
There is KDTree in GitHub - KristofferC/NearestNeighbors.jl: High performance nearest neighbor data structures and algorithms for Julia..
Thank you both!
unfortunately none of the suggestions apply…
OnlineStats offers no multidimensional adaptive tree,
while NearestNeighbors offers only static trees…
Thank you anyway
How about RegionTrees?
I just remembered that I’ve actually implemented something along those lines before
https://github.com/baggepinnen/KalmanTree.jl
I never ended up finishing it, but it could give you some hints if you’re ending up needing to implement something.
It adaptively splits the space into new cells as data arrives, and maintains a quadratic model of the data in each cell.
both packages seem useful, i will look into it thank you guys