3.5.5 把所有的东西放在一起: 面孔识别

展示使用主成分分析来降维和用执行向量机分类的面孔识别的例子。

3.5.5 把所有的东西放在一起: 面孔识别 - 图1

In [ ]:

  1. """
  2. Stripped-down version of the face recognition example by Olivier Grisel
  4. http://scikit-learn.org/dev/auto_examples/applications/face_recognition.html
  6. ## original shape of images: 50, 37
  7. """
  8. import numpy as np
  9. import pylab as pl
  10. from sklearn import cross_val, datasets, decomposition, svm
  12. # ..
  13. # .. load data ..
  14. lfw_people = datasets.fetch_lfw_people(min_faces_per_person=70, resize=0.4)
  15. perm = np.random.permutation(lfw_people.target.size)
  16. lfw_people.data = lfw_people.data[perm]
  17. lfw_people.target = lfw_people.target[perm]
  18. faces = np.reshape(lfw_people.data, (lfw_people.target.shape[0], -1))
  19. train, test = iter(cross_val.StratifiedKFold(lfw_people.target, k=4)).next()
  20. X_train, X_test = faces[train], faces[test]
  21. y_train, y_test = lfw_people.target[train], lfw_people.target[test]
  23. # ..
  24. # .. dimension reduction ..
  25. pca = decomposition.RandomizedPCA(n_components=150, whiten=True)
  26. pca.fit(X_train)
  27. X_train_pca = pca.transform(X_train)
  28. X_test_pca = pca.transform(X_test)
  30. # ..
  31. # .. classification ..
  32. clf = svm.SVC(C=5., gamma=0.001)
  33. clf.fit(X_train_pca, y_train)
  35. # ..
  36. # .. predict on new images ..
  37. for i in range(10):
  38.     print lfw_people.target_names[clf.predict(X_test_pca[i])[0]]
  39.     _ = pl.imshow(X_test[i].reshape(50, 37), cmap=pl.cm.gray)
  40.     _ = raw_input()

完整代码: faces.py