def kmeans(samples, k, cutoff):
"""
kmeans函数
"""
# 随机选k个样本点作为初始聚类中心
init_samples = random.sample(samples, k)
# 创建k个聚类，聚类的中心分别为随机初始的样本点
clusters = [Cluster([sample]) for sample in init_samples]
# 迭代循环直到聚类划分稳定
n_loop = 0
while True:
# 初始化一组空列表用于存储每个聚类内的样本点
lists = [[] for _ in clusters]
# 开始迭代
n_loop += 1
# 遍历样本集中的每个样本
for sample in samples:
# 计算样本点sample和第一个聚类中心的距离
smallest_distance = get_distance(sample, clusters[0].centroid)
# 初始化属于聚类 0
cluster_index = 0
# 计算和其他聚类中心的距离
for i in range(k - 1):
# 计算样本点sample和聚类中心的距离
distance = get_distance(sample, clusters[i + 1].centroid)
# 如果存在更小的距离，更新距离
if distance < smallest_distance:
smallest_distance = distance
cluster_index = i + 1
# 找到最近的聚类中心，更新所属聚类
lists[cluster_index].append(sample)
# 初始化最大移动距离
biggest_shift = 0.0
# 计算本次迭代中，聚类中心移动的距离
for i in range(k):
shift = clusters[i].update(lists[i])
# 记录最大移动距离
biggest_shift = max(biggest_shift, shift)
# 如果聚类中心移动的距离小于收敛阈值，即：聚类稳定
if biggest_shift < cutoff:
print("第{}次迭代后，聚类稳定。".format(n_loop))
break
# 返回聚类结果
return clusters
def kmeans(samples, k, cutoff):
"""
the function of kmeans
"""
# first ramdon choose k samples
init_samples = random.sample(samples, k)
# construct k clustering, and the initial random sample as centroid
clusters = [Cluster([sample]) for sample in init_samples]
# iterate untill reach steady state
n_loop = 0
while True:
lists = [[] for _ in clusters]
# starts
n_loop += 1
for sample in samples:
smallest_distance = get_distance(sample, clusters[0].centroid)
cluster_index = 0
for i in range(k - 1):
distance = get_distance(sample, clusters[i + 1].centroid)
if distance < smallest_distance:
smallest_distance = distance
cluster_index = i + 1
# find the centroid and update the cluster
lists[cluster_index].append(sample)
# initial the shift max distance
biggest_shift = 0.0
for i in range(k):
shift = clusters[i].update(lists[i])
# record the biggest shift distance
biggest_shift = max(biggest_shift, shift)
# if the shift distance is smaller than cutoff, the clustering stable
if biggest_shift < cutoff:
print("{} iterate, stable.".format(n_loop))
break
return clusters