K-Means：质心、分配与更新步骤
您将手动追踪 K-Means 的三次迭代，将点分配给最近的质心，重新计算质心，并在二维散点图上观察收敛过程。
K-Means：质心、分配与更新步骤 是 CoddyKit 上的免费 Machine Learning Academy 课时。 这是第 1 节课，共 4 节。 你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。 这是 Machine Learning Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Machine Learning Academy 课程共包含 4 节课。
本课时的部分内容尚未翻译，以英文显示。
What Is K-Means Clustering?
K-Means is an unsupervised algorithm that partitions n data points into k non-overlapping clusters. Unlike supervised learning, there are no labels — the algorithm discovers structure purely from the feature values. K-Means is fast, scalable, and widely used for customer segmentation, image compression, and anomaly detection.
The Three-Step Algorithm
K-Means repeats three steps until convergence: 1) Initialise — randomly place k centroids in feature space. 2) Assignment — assign every point to the nearest centroid. 3) Update — move each centroid to the mean of its assigned points. The loop stops when assignments no longer change.
Computing Distance to Centroids
In each assignment step, the Euclidean distance from every point to every centroid is computed. A point is assigned to the centroid with the smallest distance. For a point x and centroid c, the squared distance is sum((x_i - c_i)^2). Using squared distance avoids the expensive square-root and gives the same ordering.
import numpy as np
def assign_clusters(X, centroids):
# X: (n, d), centroids: (k, d)
distances = np.linalg.norm(X[:, np.newaxis] - centroids, axis=2) # (n, k)
return np.argmin(distances, axis=1) # label for each point
X = np.array([[1, 2], [3, 4], [5, 6], [8, 8]])
centroids = np.array([[2, 2], [7, 7]])
labels = assign_clusters(X, centroids)
print(labels) # [0, 0, 0, 1]The Update Step: Recomputing Centroids
After assignment, each centroid is relocated to the arithmetic mean of all points currently in its cluster. If a cluster becomes empty (no points assigned), the centroid is usually re-initialised randomly or removed. This mean-shift minimises the total within-cluster sum of squares (WCSS) — also called inertia.
import numpy as np
def update_centroids(X, labels, k):
d = X.shape[1]
new_centroids = np.zeros((k, d))
for c in range(k):
points = X[labels == c]
if len(points) > 0:
new_centroids[c] = points.mean(axis=0)
return new_centroids
X = np.array([[1, 2], [3, 4], [5, 6], [8, 8]])
labels = np.array([0, 0, 0, 1])
print(update_centroids(X, labels, k=2))Tracing Convergence by Hand
Consider four 1D points: 1, 2, 8, 9 and k=2. Init: centroids = [1, 8]. Iter 1 assignment: 1→C0, 2→C0, 8→C1, 9→C1. Iter 1 update: C0=1.5, C1=8.5. Iter 2 assignment: unchanged. Converged in 2 iterations! In higher dimensions convergence may take more steps, but the logic is identical.
Inertia: Measuring Cluster Compactness
Inertia (WCSS) is the sum of squared distances between each point and its cluster centroid. Lower inertia means tighter, more compact clusters. K-Means minimises inertia at each update step, but the algorithm is not guaranteed to find the global minimum — it can get stuck in local optima depending on initialisation.
from sklearn.cluster import KMeans
import numpy as np
X = np.array([[1, 2], [1, 4], [1, 0],
[10, 2], [10, 4], [10, 0]])
km = KMeans(n_clusters=2, random_state=42)
km.fit(X)
print('Inertia:', km.inertia_)
print('Labels:', km.labels_)
print('Centroids:', km.cluster_centers_)K-Means++ Initialisation
Random centroid initialisation often leads to slow convergence or poor local optima. K-Means++ (the scikit-learn default via init='k-means++') seeds centroids more intelligently: the first centroid is chosen randomly, and each subsequent centroid is selected with probability proportional to its squared distance from the nearest already-chosen centroid. This spreads starting points and consistently finds better solutions.
from sklearn.cluster import KMeans
import numpy as np
X = np.random.randn(300, 2)
# Default: k-means++ initialisation
km = KMeans(n_clusters=3, init='k-means++', n_init=10, random_state=0)
km.fit(X)
print('Inertia with k-means++:', round(km.inertia_, 2))
# Compare with random init
km_rand = KMeans(n_clusters=3, init='random', n_init=10, random_state=0)
km_rand.fit(X)
print('Inertia with random init:', round(km_rand.inertia_, 2))Visualising Cluster Assignments
Plotting cluster assignments on a 2D scatter shows the Voronoi partition — the decision boundaries where each point's nearest centroid changes. Plotting centroids as large stars and colouring points by cluster label makes convergence intuitive. This visualisation also reveals when clusters overlap or have unequal sizes.
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.datasets import make_blobs
X, _ = make_blobs(n_samples=300, centers=3, cluster_std=0.6, random_state=0)
km = KMeans(n_clusters=3, random_state=0)
labels = km.fit_predict(X)
plt.scatter(X[:, 0], X[:, 1], c=labels, cmap='tab10', s=30)
plt.scatter(km.cluster_centers_[:, 0], km.cluster_centers_[:, 1],
c='black', s=200, marker='*', label='Centroids')
plt.legend()
plt.title('K-Means Clusters')
plt.show()Multiple Restarts and n_init
Because K-Means can converge to local optima, scikit-learn runs the algorithm n_init times with different random seeds and keeps the result with the lowest inertia. The default is n_init=10. For small datasets 10 is usually sufficient; for large or tricky datasets you may raise it to 20 or 50. Always check the final inertia against the best-run inertia to diagnose poor convergence.
from sklearn.cluster import KMeans
import numpy as np
X = np.random.randn(500, 5)
km = KMeans(n_clusters=4, n_init=20, random_state=0)
km.fit(X)
print('Best inertia over 20 runs:', round(km.inertia_, 2))
print('Number of iterations until convergence:', km.n_iter_)Limitations of K-Means
K-Means has several well-known weaknesses: 1) Assumes spherical clusters — it struggles with elongated or crescent shapes. 2) Sensitive to outliers — a distant outlier pulls the centroid away from the true cluster mean. 3) Requires k upfront — you must know or estimate the number of clusters before fitting. 4) Feature scale matters — always standardise features before running K-Means so large-scale variables do not dominate distances.
Running K-Means with scikit-learn
In practice, using sklearn.cluster.KMeans is the standard approach. Key parameters: n_clusters (k), init (default 'k-means++'), n_init, max_iter (default 300), and random_state. After fitting, km.labels_ contains cluster assignments, km.cluster_centers_ holds centroid positions, and km.inertia_ reports WCSS.
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.datasets import load_iris
X, _ = load_iris(return_X_y=True)
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)
km = KMeans(n_clusters=3, random_state=42)
km.fit(X_scaled)
print('Cluster sizes:', {i: (km.labels_ == i).sum() for i in range(3)})
print('Inertia:', round(km.inertia_, 2))Quick Check
Test your understanding of K-Means clustering concepts from this lesson.
Lesson Recap
In this lesson you learned: K-Means iterates assignment and update steps until cluster memberships stabilise, inertia (WCSS) measures compactness and is minimised by each update, and K-Means++ initialisation and multiple restarts help avoid poor local optima. Next up we explore how to choose the right value of k using the elbow method and silhouette score.
用 AI 导师学习 Python — 免费
在浏览器中编写并运行真实代码，获得全天候 AI 导师的即时帮助，并在网页或应用中继续学习。
- 课程
- 30
- 课程
- 120
常见问题解答
「K-Means：质心、分配与更新步骤」课时是免费的吗？
是的 — 「K-Means：质心、分配与更新步骤」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Machine Learning Academy 课程的其余内容，请升级到 CoddyKit PRO。 Machine Learning Academy 课程共包含 4 节课。
「K-Means：质心、分配与更新步骤」这节课中我会学到什么？
您将手动追踪 K-Means 的三次迭代，将点分配给最近的质心，重新计算质心，并在二维散点图上观察收敛过程。 你通过在浏览器中直接运行的动手代码来练习 Machine Learning Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。
学习 Machine Learning Academy 需要有经验吗？
无需任何先前经验。CoddyKit 上的 Machine Learning Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。 这是第 1 节课，共 4 节。
「K-Means：质心、分配与更新步骤」课时需要多长时间？
大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。
我能在这节 Machine Learning Academy 课中编写并运行代码吗？
能。每节 Machine Learning Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。
此课程中的所有课时
- K-Means：质心、分配与更新步骤
- 选择 K：肘部法与轮廓系数
- DBSCAN：核心点、边界点与噪声点
- 用于客户细分的聚类：端到端示例