Python：给定数据集计算样本之间的距离矩阵

            
              import numpy as np
from sklearn.datasets import load_iris

iris = load_iris()
# data = iris.data
# print(data[0])
# print(data[2])
# print(type(iris.data))
# print(iris.data.shape)
# LenRow, LenColumn = iris.data.shape
# print("LenRow={}".format(LenRow))
# print("LenColumn={}".format(LenColumn))


def Distance(datarow_1,datarow_2):
    a = len(datarow_1)
    b = len(datarow_2)
    distance = 0
    term = 0
    if a != b:
        return None
    else:
        for i in range(0,a):
            term += (datarow_1[i] - datarow_2[i]) ** 2
        distance = np.sqrt(term)
    return distance

def DistanceMatrix(data):
    LenRow, LenColumn = data.shape
    Dis_Mat = np.zeros((LenRow,LenRow))
    for i in range(0,LenRow):
        for j in range(0,LenRow):
            if i < j:
                Dis_Mat[i,j]=Dis_Mat[j,i] = Distance(data[i],data[j])

    return Dis_Mat

if __name__ == "__main__":
    # data = iris.data
    # dis = Distance(data[0],data[3])
    # print(dis)
    Dis_Mat = DistanceMatrix(iris.data)
    print(type(Dis_Mat))
    print(Dis_Mat[2,3])
    print(Dis_Mat[3,2])
    print(Dis_Mat[4,5])
    print(Dis_Mat[5,4])
    print(Dis_Mat[1,1])
    print(Dis_Mat[12,12])

            
          

给聚类算法做准备