相似性度量方法（距离）总结

在做分类时常常需要估算不同样本之间的相似性度量，这时通常采用的方法就是计算样本间的距离。采用什么样的方法计算距离关系到分类的正确与否。

距离的python实现

一、欧式距离

# -*- coding: utf-8 -*-
import numpy as np

def eucliDist(A,B):
    return np.sqrt(sum(np.power((A-B),2)))

X=np.array([1,2,3,4])
Y=np.array([0,1,2,3])
print('Euclidean distance:',eucliDist(X,Y))

二、曼哈顿距离

# -*- coding: utf-8 -*-
import numpy as np

def manhaDist(A,B):
    return np.sum(np.abs(A-B))

X=np.array([1,2,3])
Y=np.array([4,5,6])

print('Manhattan distance:',manhaDist(X,Y))

三、切比雪夫距离

# -*- coding: utf-8 -*-
import numpy as np

def chebyDist(A,B):
    return np.max(np.abs(A-B))

X=np.array([1,2,3])
Y=np.array([4,7,5])

print('Chebyshev distance:',chebyDist(X,Y))

四、闵可夫斯基距离

# -*- coding: utf-8 -*-
from math import *
from decimal import Decimal

def nth_root(value, n_root):
    root_value = 1 / float(n_root)
    return round(Decimal(value) ** Decimal(root_value), 3)

def minkowski_distance(x, y, p_value):
    return nth_root(sum(pow(abs(a - b), p_value) for a, b in zip(x, y)), p_value)

print(minkowski_distance([0, 3, 4, 5], [7, 6, 3, -1], 3))

五、标准化欧式距离

# -*- coding: utf-8 -*-
import numpy as np

def standEuDist(A,B):
    T=np.vstack([A,B])
    sk=np.var(T,axis=0,ddof=1)
    return np.sqrt(((A-B)**2/sk).sum())

X=np.array([1,2,3,4])
Y=np.array([0,1,2,3])

print('Standardized Euclidean distance:',standEuDist(X,Y))

六、余弦距离

# -*- coding: utf-8 -*-
import numpy as np
from scipy.spatial.distance import pdist

def cosDist(A,B):
    S=np.vstack([A,B])
    dist=1-pdist(S,'cosine')
    return str(dist)

X = [1, 2, 3, 4]
Y = [5, 6, 7, 8]

print("Cosine distance:",cosDist(X,Y))

七、汉明距离

# -*- coding: utf-8 -*-
from numpy import *

def hamDist(matV):
    smstr=nonzero(matV[0]-matV[1])
    return (shape(smstr[0])[0])

matV = mat([[1,0,0,1,0,1,0,0,1],[0,1,1,0,0,0,1,1,1]])

print('Hamming distance:',hamDist(matV))

八、杰卡德距离

# -*- coding: utf-8 -*-
import scipy.spatial.distance as dist
import numpy as np

def jacDist(A,B):
    matv=np.array([A,B])
    return dist.pdist(matv,'jaccard')

X=np.array([1,1,0,1,0,1,0,0,1])
Y=np.array([0,1,1,0,0,0,1,1,1])

print('Jaccard distance:',jacDist(X,Y))

九、马氏距离

# -*- coding: utf-8 -*-
from numpy import *
import numpy as np

def mahaDist(x):
    xT=x.T
    print(xT)
    D=np.cov(xT)
    invD=np.linalg.inv(D)
    print(invD)
    tp=x[0]-x[1]
    return np.sqrt(dot(dot(tp,invD),tp.T))
x = np.array([[3,4],[5,6],[2,2],[8,4]])
print('Mahalanobies distance:',mahaDist(x))