快捷導(dǎo)航

使用python實(shí)現(xiàn)3D聚類圖示例代碼

更新時間：2024年08月20日 10:47:06 作者：兜里沒有一毛錢

這篇文章主要介紹了使用python實(shí)現(xiàn)3D聚類圖效果,本文通過實(shí)例代碼給大家介紹的非常詳細(xì),感興趣的朋友跟隨小編一起看看吧

一、以實(shí)現(xiàn)三個類別聚類為例

代碼：

import pandas as pd
import numpy as np
from sklearn.decomposition import PCA
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.impute import SimpleImputer
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
# 讀取數(shù)據(jù)
data = pd.read_csv('E:\\shujuji\\Goods\\man.csv')
# 選擇用于聚類的列
features = ['Weight', 'BMI', 'Lung Capacity Score', '50m Running Score',   
            'Standing Long Jump Score', 'Sitting Forward Bend Score',   
            '1000m Running Score', 'Pulling Up Score', 'Total Score']
X = data[features]
# 處理缺失值
imputer = SimpleImputer(strategy='mean')
X_imputed = imputer.fit_transform(X)
# 數(shù)據(jù)標(biāo)準(zhǔn)化
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X_imputed)
# 應(yīng)用PCA降維到3維
pca = PCA(n_components=3)
X_pca = pca.fit_transform(X_scaled)
# 執(zhí)行K-means聚類
# 假設(shè)我們想要3個聚類
kmeans = KMeans(n_clusters=9, random_state=0).fit(X_pca)
labels = kmeans.labels_
# 將聚類標(biāo)簽添加到原始DataFrame中
data['Cluster'] = labels
# 3D可視化聚類結(jié)果
fig = plt.figure(1, figsize=(8, 6))
ax = fig.add_subplot(111, projection='3d')
unique_labels = set(labels)
colors = ['r', 'g', 'b']
for k, c in zip(unique_labels, colors):
    class_member_mask = (labels == k)
    xy = X_pca[class_member_mask]
    ax.scatter(xy[:, 0], xy[:, 1], xy[:, 2], c=c, label=f'Cluster {k}')
ax.set_title('PCA of Fitness Data with K-means Clustering')
ax.set_xlabel('Principal Component 1')
ax.set_ylabel('Principal Component 2')
ax.set_zlabel('Principal Component 3')
plt.legend()
plt.show()
# 打印每個聚類的名稱和對應(yīng)的數(shù)據(jù)點(diǎn)數(shù)量
cluster_centers = kmeans.cluster_centers_
for i in range(3):
    cluster_data = data[data['Cluster'] == i]
    print(f"Cluster {i}: Count: {len(cluster_data)}")
# 評估聚類效果
from sklearn import metrics
print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X_pca, labels))

實(shí)現(xiàn)效果：

二、實(shí)現(xiàn)3個聚類以上，以9個類別聚類為例

import pandas as pd
import numpy as np
from sklearn.decomposition import PCA
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.impute import SimpleImputer
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
# 讀取數(shù)據(jù)
data = pd.read_csv('E:\\shujuji\\Goods\\man.csv')
# 選擇用于聚類的列
features = ['Weight', 'BMI', 'Lung Capacity Score', '50m Running Score',   
            'Standing Long Jump Score', 'Sitting Forward Bend Score',   
            '1000m Running Score', 'Pulling Up Score', 'Total Score']
X = data[features]
# 處理缺失值
imputer = SimpleImputer(strategy='mean')
X_imputed = imputer.fit_transform(X)
# 數(shù)據(jù)標(biāo)準(zhǔn)化
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X_imputed)
# 應(yīng)用PCA降維到3維
pca = PCA(n_components=3)
X_pca = pca.fit_transform(X_scaled)
# 執(zhí)行K-means聚類
# 假設(shè)我們想要9個聚類
kmeans = KMeans(n_clusters=9, random_state=0).fit(X_pca)
labels = kmeans.labels_
# 將聚類標(biāo)簽添加到原始DataFrame中
data['Cluster'] = labels
# 3D可視化聚類結(jié)果
fig = plt.figure(1, figsize=(8, 6))
ax = fig.add_subplot(111, projection='3d')
unique_labels = set(labels)
colors = ['r', 'g', 'b', 'c', 'm', 'y', 'k', 'orange', 'purple']
for k, c in zip(unique_labels, colors):
    class_member_mask = (labels == k)
    xy = X_pca[class_member_mask]
    ax.scatter(xy[:, 0], xy[:, 1], xy[:, 2], c=c, label=f'Cluster {k}')
ax.set_title('PCA of Fitness Data with K-means Clustering')
ax.set_xlabel('Principal Component 1')
ax.set_ylabel('Principal Component 2')
ax.set_zlabel('Principal Component 3')
plt.legend()
plt.show()
# 打印每個聚類的名稱和對應(yīng)的數(shù)據(jù)點(diǎn)數(shù)量
cluster_centers = kmeans.cluster_centers_
for i in range(9):
    cluster_data = data[data['Cluster'] == i]
    print(f"Cluster {i}: Count: {len(cluster_data)}")
# 評估聚類效果
from sklearn import metrics
print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X_pca, labels))

實(shí)現(xiàn)效果;