import numpy as npdef shouldStop(oldCentroids, centroids, iterations, maxIt):"""
    :param oldCentroids: 老的聚类中心
    :param centroids: 聚类中心
    :param iterations: 第几次迭代
    :param maxIt: 最大迭代次数
    :return: True or False
    """if iterations > maxIt:return Truereturn np.array_equal(oldCentroids, centroids)def updateLabels(dataSet, centroids):"""
    更新数据集中每个点的类别
    :param dataSet: 数据集
    :param centroids: 聚类中心
    """numPoints, numDim = dataSet.shapefor i in range(0, numPoints):dataSet[i, -1] = getLabelFromClosestCentroid(dataSet[i, :-1], centroids)def getLabelFromClosestCentroid(dataSetRow, centroids):"""
    :param dataSetRow:数据集中一个实例（一个点）
    :param centroids: 聚类中心
    :return: 返回属于哪个标签（属于哪个类）
    """label = centroids[0, -1]  # 初始化类别为第0个实例的类别minDist = np.linalg.norm(dataSetRow - centroids[0, :-1])  # 初始化最小距离为与第0个聚类中心的距离for i in range(1, centroids.shape[0]):dist = np.linalg.norm(dataSetRow - centroids[i, :-1])  # 欧式距离if dist < minDist:minDist = distlabel = centroids[i, -1]print("minDist:", minDist)return labeldef getCentroids(dataSet, k):"""
    # Each centroid is the geometric mean of the points that
    # have that centroid's label. Important: If a centroid is empty (no points have
    # that centroid's label) you should randomly re-initialize it.
    :param dataSet: 数据集
    :param k: 几个类别
    :return: 更新之后的聚类中心
    """result = np.zeros((k, dataSet.shape[1]))for i in range(1, k + 1):oneCluster = dataSet[dataSet[:, -1] == i, :-1]result[i - 1, :-1] = np.mean(oneCluster + np.spacing(1), axis=0)  # 防止对空列表求平均result[i - 1, -1] = ireturn result

def kmeans(X, k, maxIt):numPoints, numDim = X.shapedataSet = np.zeros((numPoints, numDim + 1))dataSet[:, :-1] = Xcentroids = dataSet[np.random.randint(numPoints, size=k), :]# centroids = dataSet[0:2, :]         # Randomly assign labels to initial centorid给初始中心随机分配标签centroids[:, -1] = range(1, k + 1)iterations = 0oldCentroids = Nonewhile not shouldStop(oldCentroids, centroids, iterations, maxIt):print("iteration: \n", iterations)print("dataSet: \n", dataSet)print("centroids: \n", centroids)# Save old centroids for convergence test. Book keeping.oldCentroids = np.copy(centroids)iterations += 1updateLabels(dataSet, centroids)centroids = getCentroids(dataSet, k)return dataSet

iteration: 0
dataSet: [[1. 1. 0.][2. 1. 0.][4. 3. 0.][5. 4. 0.]]
centroids: [[1. 1. 1.][4. 3. 2.]]
minDist: 0.0
minDist: 1.0
minDist: 0.0
minDist: 1.4142135623730951
iteration: 1
dataSet: [[1. 1. 1.][2. 1. 1.][4. 3. 2.][5. 4. 2.]]
centroids: [[1.5 1.  1. ][4.5 3.5 2. ]]
minDist: 0.5
minDist: 0.5
minDist: 0.7071067811865476
minDist: 0.7071067811865476
final result:
[[1. 1. 1.][2. 1. 1.][4. 3. 2.][5. 4. 2.]]