kdtree library
[dependencies]
kdtree = { git = "https://github.com/bave/kdtree.git" }/src/bin/main.rs
extern crate kdtree;
//use self::kdtree::*;
use kdtree::*;
#[derive(Debug)]
struct Point
{
id: String,
vec: [f64;Point::DIM],
}
impl TraitPoint for Point
{
const DIM: usize = 2;
#[inline]
fn dim(&self, depth: usize) -> f64
{
let axis = depth % Point::DIM;
self.vec[axis]
}
}
fn main()
{
let mut vec_point : Vec<Point> = Vec::new();
vec_point.push(Point{id: "hoge1".to_string(), vec: [1.0, 2.0]});
vec_point.push(Point{id: "hoge2".to_string(), vec: [3.0, 4.0]});
vec_point.push(Point{id: "hoge3".to_string(), vec: [5.0, 6.0]});
let mut kdt = KDTree::new(&vec_point);
kdt.build();
print!("{:#?}\n", kdt);
let p = Point{id: "query".to_string(), vec: [3.1, 4.1]};
let k = kdt.knn_search(&p, 1);
print!("knn return indices: {:?}\n", k);
print!("nearlest neighbor Point:{:?}\n", vec_point[k[0]]);
let r = kdt.radius_search(&p, 1.0);
print!("radius search return indices: {:?}\n", r);
print!("nearlest neighbor Point:{:?}\n", vec_point[r[0]]);
}$ cargo run
KDTree {
root: Some(
Node {
idx: 1,
axis: 0,
left: Some(
Node {
idx: 0,
axis: 1,
left: None,
right: None,
},
),
right: Some(
Node {
idx: 2,
axis: 1,
left: None,
right: None,
},
),
},
),
points: [
Point {
id: "hoge1",
vec: [
1.0,
2.0,
],
},
Point {
id: "hoge2",
vec: [
3.0,
4.0,
],
},
Point {
id: "hoge3",
vec: [
5.0,
6.0,
],
},
],
}
knn return indices: [1]
nearlest neighbor Point:Point { id: "hoge2", vec: [3.0, 4.0] }
radius search return indices: [1]
nearlest neighbor Point:Point { id: "hoge2", vec: [3.0, 4.0] }