import numpy as np class KMeans: def __init__(self, k=2, tolerance=0.0001, max_iterations=300): self.k = k self.tolerance = tolerance self.max_iterations = max_iterations def fit(self, data): self.centroids = {} # Initialize the centroids, the first 'k' data points in the dataset for i in range(self.k): self.centroids[i] = data[i] # Begin the iterations for i in range(self.max_iterations): self.classes = {} for j in range(self.k): self.classes[j] = [] # Find the distance between the point and cluster; choose the nearest centroid for point in data: distances = [np.linalg.norm(point - self.centroids[centroid]) for centroid in self.centroids] classification = distances.index(min(distances)) self.classes[classification].append(point) previous = dict(self.centroids) # Calculate the mean of the clusters to update the centroids for classification in self.classes: self.centroids[classification] = np.average(self.classes[classification], axis=0) # Check if converged is_converged = True for centroid in self.centroids: original_centroid = previous[centroid] curr_centroid = self.centroids[centroid] if np.sum((curr_centroid - original_centroid) / original_centroid * 100.0) > self.tolerance: is_converged = False # If converged, break out of the loop if is_converged: break解释具体代码含义

这段代码是 Python 语言中使用 NumPy 库实现的 K 均值聚类算法。`import numpy as np` 是导入 NumPy 库并将其重命名为 `np`，方便在代码中使用。`class KMeans` 是定义了一个 KMeans 类，用于实现 K 均值聚类算法。`__init__` 函数是类的构造函数，用于初始化类的属性，包括聚类数目 `k`，容忍度 `tolerance` 和最大迭代次数 `max_iterations`。`fit` 函数是实现 K 均值聚类算法的主要函数，会根据输入的数据集 `data` 进行迭代，计算出聚类中心点 `centroids`，直到达到容忍度或者最大迭代次数的限制。