SVM算法识别政治家面部识别

实验名称： SVM算法识别政治家面部识别

一、问题描述

国际政治家面部识别。在此数据集上，采用SVM算法实现：
给定以张图片，预测图片中的任务是否是这位政治家。

二、实验目的

学习SVM算法的算法。

三、实验内容

数据导入

from sklearn.datasets import fetch_lfw_people

数据预处理

lfw_people = fetch_lfw_people(min_faces_per_person=70, resize=0.4)
X = lfw_people.data
y = 2 * (lfw_people.target==3).astype(np.int32).reshape(-1,1) - 1

算法描述

当线性不可分时，即不存在分离平面，软间隔SVM算法仍然能给出一个近似的分类平面。

class SoftSVM(SVM):
    def __init__(self, C = 1000):
        self.C = C
    
    def get_H(self, Lambda, i,j, y):
        C = self.C
        if y[i]==y[j]:
            return min(C, Lambda[i] + Lambda[j])
        else:
            return min(C, C + Lambda[j] - Lambda[i]) 
    
    def get_L(self, Lambda, i, j, y):
        if y[i]==y[j]:
            return max(0, Lambda[i] + Lambda[j] - self.C)
        else:
            return max(0, Lambda[j] - Lambda[i])

主要代码

import numpy as np
from sklearn.datasets import fetch_lfw_people
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score

faces = fetch_lfw_people(min_faces_per_person=70, resize=0.4)
print(faces.target_names)
print(faces.images.shape)


lfw_people = fetch_lfw_people(min_faces_per_person=70, resize=0.4)
X = lfw_people.data
y = 2 * (lfw_people.target==3).astype(np.int32).reshape(-1,1) - 1
# print(lfw_people.target)
m, h, w = lfw_people.images.shape
print(h,w)
# plt.imshow(X[1].reshape(h, w))
plt.imshow(X[3].reshape(h, w))
plt.show()

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)
model = SVC(kernel='rbf', C=1, gamma=1)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print("accuracy = {}".format(accuracy))

四、实验结果及分析

五、遇到的问题和解决方法

六、获得资源

https://download.csdn.net/download/m0_61504367/85320352