#導(dǎo)入需要的包
import numpy as np
from sklearn.cluster import KMeans
from sklearn.datasets import make_moons#生成數(shù)據(jù)集，形成兩個(gè)彎月。
from sklearn.datasets import make_circles#生成數(shù)據(jù)集，形成一個(gè)二維的大圓，包含一個(gè)小圓
from sklearn.datasets import make_blobs#為聚類生成符合正態(tài)分布的數(shù)據(jù)集，產(chǎn)生一個(gè)數(shù)據(jù)集和相應(yīng)的標(biāo)簽
import matplotlib.pyplot as plt

def get_eigen(L, num_clusters):#獲取特征
    eigenvalues, eigenvectors = np.linalg.eigh(L)#獲取特征值  特征向量
    best_eigenvalues = np.argsort(eigenvalues)[0:num_clusters]#argsort函數(shù)返回的是數(shù)組值從小到大的索引值
    U = np.zeros((L.shape[0], num_clusters))
    U = eigenvectors[:, best_eigenvalues]#將這些特征取出 構(gòu)成新矩陣
    return U

#K-Means聚類
def cluster(data, num_clusters):
    data = np.array(data)
    W = affinity_matrix(data)
    D = getD(W)
    L = getL(D, W)
    eigenvectors = get_eigen(L, num_clusters)
    clf = KMeans(n_clusters=num_clusters)
    s = clf.fit(eigenvectors)  # 聚類
    label = s.labels_
    return label

#導(dǎo)入需要的包
import numpy as np
from sklearn.cluster import KMeans
from sklearn.datasets import make_moons#生成數(shù)據(jù)集，形成兩個(gè)彎月。
from sklearn.datasets import make_circles#生成數(shù)據(jù)集，形成一個(gè)二維的大圓，包含一個(gè)小圓
from sklearn.datasets import make_blobs#為聚類生成符合正態(tài)分布的數(shù)據(jù)集，產(chǎn)生一個(gè)數(shù)據(jù)集和相應(yīng)的標(biāo)簽
import matplotlib.pyplot as plt

#定義高斯核函數(shù)
def kernel(x1, x2, sigma_sq=0.05):
    return np.exp(-(np.linalg.norm(x1 - x2) ** 2) / (2 * sigma_sq ** 2))

#定義相似度矩陣
def affinity_matrix(X):
    A = np.zeros((len(X), len(X)))#零矩陣
    for i in range(len(X) - 1):#長(zhǎng)度為len(x) 但是從0開始
        for j in range(i + 1, len(X)):#從1開始，到len(x) 是方陣 為啥下角標(biāo)取值的初始值不同？？？
            A[i, j] = A[j, i] = kernel(X[i], X[j])
    return A#通過(guò)高斯核的計(jì)算 給矩陣賦予新值    10*10

# 計(jì)算度矩陣
def getD(A):
    D = np.zeros(A.shape)
    for i in range(A.shape[0]):
        D[i, i] = np.sum(A[i, :])
    return D

#計(jì)算拉普拉斯矩陣
def getL(D, A):
    L = D - A
    return L


def get_eigen(L, num_clusters):#獲取特征
    eigenvalues, eigenvectors = np.linalg.eigh(L)#獲取特征值  特征向量
    best_eigenvalues = np.argsort(eigenvalues)[0:num_clusters]#argsort函數(shù)返回的是數(shù)組值從小到大的索引值
    U = np.zeros((L.shape[0], num_clusters))
    U = eigenvectors[:, best_eigenvalues]#將這些特征取出 構(gòu)成新矩陣
    return U

#K-Means聚類
def cluster(data, num_clusters):
    data = np.array(data)
    W = affinity_matrix(data)
    D = getD(W)
    L = getL(D, W)
    eigenvectors = get_eigen(L, num_clusters)
    clf = KMeans(n_clusters=num_clusters)
    s = clf.fit(eigenvectors)  # 聚類
    label = s.labels_
    return label


def plotRes(data, clusterResult, clusterNum):
    """
    結(jié)果可似化
    : data:  樣本集
    : clusterResult: 聚類結(jié)果
    : clusterNum: 聚類個(gè)數(shù)
    :return:
    n = len(data)
    scatterColors = ['black', 'blue', 'red', 'yellow', 'green', 'purple', 'orange']
    for i in range(clusterNum):
        color = scatterColors[i % len(scatterColors)]
        x1 = []
        y1 = []
        for j in range(n):
            if clusterResult[j] == i:
                x1.append(data[j, 0])
                y1.append(data[j, 1])
        plt.scatter(x1, y1, c=color, marker='+')


if __name__ == '__main__':
# # #月牙形數(shù)據(jù)集,sigma=0.1
# #     # cluster_num = 2
# #     # data, target = make_moons()
# #     # label = cluster(data, cluster_num)
# #     # print(label)
# #     # plotRes(data, label, cluster_num)
# #
#     # 圓形數(shù)據(jù)集,sigma=0.05
    cluster_num = 2
    data, target = make_circles(n_samples=1000, shuffle=True, noise=0.05, factor=0.5)
    label = cluster(data, cluster_num)
    print(label)
    plotRes(data, label, cluster_num)
# #    #  # 正態(tài)數(shù)據(jù)集
# #    # # n_samples是待生成的樣本的總數(shù)。
# #    #  # n_features是每個(gè)樣本的特征數(shù)。
# #    #  # centers表示類別數(shù)。
# #    #  # cluster_std表示每個(gè)類別的方差，例如我們希望生成2類數(shù)據(jù)，其中一類比另一類具有更大的方差，可以將cluster_std設(shè)置為[1.0, 3.0]。
# #    #  cluster_num = 2
# #    #  data, target = make_blobs(n_samples=1500, n_features=2, centers=4, random_state=24)
# #    #  label = cluster(data, cluster_num)
# #    #  print(label)
# #    #  plt.subplot(121)
# #    #  plotRes(data, target, cluster_num)
# #    #  plt.subplot(122)
# #    #  plotRes(data, label, cluster_num)
plt.show()