人工智能领域机器学习算法支持向量机python代码

# -*- coding: utf-8 -*- """ Created on Tue Mar 9 10:57:18 2021 @author: lenovo """ import numpy as np import pandas as pd from matplotlib import pyplot as plt #读取csv 数据 df = pd.read_csv("samples.txt",header=None,names=["x","y","label"]) print(df.head()) #分片选择样本实例，标记 X,y = df.iloc[:,:-1].values,df["label"].values X0 = X[y==0] X1 = X[y==1] #画出样本点 #设置显示中文 plt.rcParams["font.sans-serif"] = ["SimHei"] plt.rcParams["axes.unicode_minus"] = False #训练SVM分类模型 from sklearn.svm import SVC #使用不同的核函数，实例化模型对象 linear_model = SVC(C=1.0,kernel="linear") poly_model = SVC(C=1.0,kernel="poly",degree=3) rbf1_model = SVC(C=1.0,kernel="rbf",gamma=5) rbf2_model = SVC(C=1.0,kernel="rbf",gamma=0.1) models = [linear_model,poly_model,rbf1_model,rbf2_model] titles = ["线性核","三阶多项式核","高斯核$\gamma = 5$","高斯核$\gamma = 0.1$"] #生成网格点 x1_min,x2_min = X.min(axis=0) x1_max,x2_max =X.max(axis=0) x1_range = np.linspace(x1_min-1,x1_max+1,100) x2_range = np.linspace(x2_min-1,x2_max+1,100) #张成网格 xx1,xx2 = np.meshgrid(x1_range,x2_range) #组织成样本点 xx = np.c_[xx1.ravel(),xx2.ravel()] #遍历模型，训练，画出决策线 plt.figure("decision line",facecolor="snow",figsize=(10,8)) for idx,model in enumerate(models): model.fit(X,y) #预测网格点 y_pred = model.predict(xx) #画子图 plt.subplot(2,2,idx+1) plt.subplots_adjust(left=0.1,right=1,top=1,bottom=0.1,wspace=0.2,hspace=0.3) #画点 plt.scatter(X0[:,0],X0[:,1],s=80,c="r",edgecolors=None,marker="s",label="0类") plt.scatter(X1[:,0],X1[:,1],s=80,c="b",edgecolors=None,marker="^",label="1类") plt.title(titles[idx],fontsize=20) plt.xlabel("x 轴",fontsize=20) plt.ylabel("y 轴",fontsize=20) plt.grid(linestyle="-") plt.legend(loc="upper right",frameon=True,framealpha=0.3) #画决策线 plt.contour(xx1,xx2,y_pred.reshape(xx1.shape),cmap="cool") plt.show() #可以更改 C gamma值观察 决策线的变化