import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_blobs
x,y = make_blobs(
    n_samples=40,centers=2,random_state=0
)
plt.scatter(x[:,0],x[:,1],c=y)
plt.show()

from sklearn.inspection import DecisionBoundaryDisplay

def plot_decision_boundary(x, y, clf):
    """
    绘制分类模型的决策边界

    参数:
    x (array-like): 特征数据，形状为 (n_samples, n_features)
    y (array-like): 标签数据，形状为 (n_samples,)
    clf (estimator): 训练好的sklearn分类模型
    """
    # 创建图形
    fig, ax = plt.subplots(figsize=(10, 6))

    # 绘制决策边界
    disp = DecisionBoundaryDisplay.from_estimator(
        clf,
        x,
        response_method="predict",
        alpha=0.5,
        ax=ax,
        grid_resolution=300,
        cmap=plt.cm.coolwarm
    )

    # 绘制训练点
    scatter = ax.scatter(
        x[:, 0], x[:, 1], c=y, edgecolor="k",
        alpha=0.8, cmap=plt.cm.coolwarm, s=50
    )

    # 添加图例
    legend1 = ax.legend(*scatter.legend_elements(), title="Classes")
    ax.add_artist(legend1)

    # 设置标题和坐标轴标签
    ax.set_title(f"Decision Boundary of {type(clf).__name__}")
    ax.set_xlabel("Feature 1")
    ax.set_ylabel("Feature 2")

    plt.tight_layout()
    plt.show()

from sklearn.svm import LinearSVC
clf = LinearSVC(C=1)
clf.fit(x,y)
plot_decision_boundary(x,y,clf)

运行结果

 c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( 

def plot_svm_margin(x,y,clf,ax=None):
    from sklearn.inspection import DecisionBoundaryDisplay
    DecisionBoundaryDisplay.from_estimator(clf, x, ax=ax,plot_method='contour',colors='k',alpha=0.5,levels=[-1,0,1])
    plt.scatter(x[:,0],x[:,1],c=y)

plot_svm_margin(x,y,clf)
plt.show()

plt.rcParams['figure.figsize'] = (12,8)
params = [0.1,1,10,100]
for i,c in enumerate(params):
    clf = LinearSVC(C=c,random_state=0)
    clf.fit(x,y)
    ax = plt.subplot(2,2,i+1)
    plt.title("C = "+str(c))
    plot_svm_margin(x,y,clf,ax)
plt.show()

运行结果

 c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( 

我们可以看到，随着C增大，容错空间一直在减小

多分类问题

from sklearn import datasets
iris = datasets.load_iris()
X = iris.data
y = iris.target
X.shape, y.shape

运行结果

 ((150, 4), (150,)) 

clf = LinearSVC(C=0.1,penalty='l2',multi_class='ovr') # 默认不支持OvO策略
clf.fit(X,y)

运行结果

 c:\Users\32980\AppData\Local\Programs\Python\Python38\lib\site-packages\sklearn\svm\_classes.py:32: FutureWarning: The default value of `dual` will change from `True` to `'auto'` in 1.5. Set the value of `dual` explicitly to suppress the warning. warnings.warn( 

运行结果

 LinearSVC(C=0.1) 

clf.score(X,y)

运行结果

 0.96