flann-lib / flann

Fast Library for Approximate Nearest Neighbors
http://people.cs.ubc.ca/~mariusm/flann
Other
2.21k stars 646 forks source link

spatial search on large data set #487

Open alisheikholeslam opened 2 years ago

alisheikholeslam commented 2 years ago

I am seeking for pythonic libraries that could be used for spatial searching with good performance using large data set. I am very eager to know if this library can satisfy my needs in terms of execution runtimes and performance.

My problem is:

I have lots of spheres and want to find corresponding spheres in a distance of each sphere as fast as it is possible because I must call this function many times during my simulation. I have written a python code using numpy and scipy in one of issues in stack overflow. But, as it is mentioned in that issue, it is not appropriate for large data set. Is there any prepared example code or any done time-consuming evaluations to investigate capability of this library on this subject? Does this library can perform this spatial searching for 100 nearest spheres on 500000 spheres less than a second or around it? How much time will be your estimation on this subject?

I would appreciate any explanation or for referring to any prepared code to evaluate runtime by this library.