from matplotlib import pyplot as plt from sklearn.datasets import fetch_lfw_people from sklearn.svm import SVC from sklearn.decomposition import PCA from sklearn.pipeline import make_pipeline from sklearn.model_selection import train_test_split from sklearn.model_selection import GridSearchCV import time
faces=fetch_lfw_people(min_faces_per_person=60) #一个人人脸图像数少于60就不作为样本
fig,ax=plt.subplots(3,5) for i, axi in enumerate(ax.flat): axi.imshow(faces.images[i],cmap=“bone”) axi.set(xticks=[],yticks=[],xlabel=faces.target_names[faces.target[i]])
#每个图62*47，需要降维,降到150维
pca=PCA(n_components=150,whiten=True,random_state=42) #random_state就是随机种子，控制每次随机结果一样 svc=SVC(kernel=“rbf”,class_weight=“balanced”) model=make_pipeline(pca,svc)
#在sklearn中构造训练集和验证集 Xtrain,Xtest,ytrain,ytest=train_test_split(faces.data,faces.target,random_state=40)
#用GridSearchCV来寻找参数，c,gamma,给出几个值，遍历 param_grid={“svc__C”:[2,3,4], “svc__gamma”:[0.0015,0.0017,0.0019]
} grid=GridSearchCV(model,param_grid) %time grid.fit(Xtrain,ytrain) print(grid.best_params_)
model=grid.best_estimator_ yfit=model.predict(Xtest) yfit.shape
#画出结果 fig, ax=plt.subplots(4,6) for i, axi in enumerate(ax.flat): axi.imshow(Xtest[i].reshape(62,47),cmap=“bone”) axi.set(xticks=[],yticks=[]) axi.set_ylabel(faces.target_names[yfit[i]].split()[-1], color=“black” if yfit[i]==ytest[i] else “red” ) fig.suptitle(“Predicted names;Incorrect labels in red”,size=18)
