球树结构

1 基本使用方法

sklearn.neighbors.BallTree(X, leaf_size=40, metric='minkowski', **kwargs)

2 主要参数说明

3 举例

3.1 最近的K个邻居

import numpy as np
from sklearn.neighbors import BallTree
X = np.random.randn(10,3)
tree = BallTree(X, leaf_size=2)              
dist, ind = tree.query(X[:2], k=3)                print(ind)  
# 最近的k个邻居的index
'''
[[0 4 5][1 2 8]]
'''print(dist)  
# 最近的k个邻居的距离
'''
[[0.         0.86677441 1.16406937][0.         0.95190704 1.32997164]]
'''

3.2 位于查询点指定半径内的所有邻居点

• 查找位于每个查询点半径内的点的索引

query_radius(X, r, return_distance=False, count_only=False, sort_results=False)

import numpy as np
from sklearn.neighbors import BallTree
X = np.random.randn(10,3)
tree = BallTree(X, leaf_size=2)              tree.query_radius(X[:2],r=[0.1,5])
'''
array([array([0], dtype=int64),array([6, 2, 4, 7, 8, 1, 0, 3, 9, 5], dtype=int64)], dtype=object)
'''

import numpy as np
from sklearn.neighbors import BallTree
X = np.random.randn(10,3)
tree = BallTree(X, leaf_size=2)              tree.query_radius(X[:2],r=[0.1,5],return_distance=True)
'''
(array([array([0], dtype=int64),array([8, 2, 3, 6, 1, 9, 7, 0, 5, 4], dtype=int64)], dtype=object),array([array([0.]),array([2.18948629, 1.05002031, 1.48036256, 1.54854719, 0.        ,2.37799982, 3.36371823, 2.63138373, 2.54630893, 3.57322436])],dtype=object))
'''