PCA的python实现

import numpy as np
from sklearn.decomposition import PCA
import matplotlib.pyplot as plt

iris_data = np.loadtxt('iris_train_data.csv', skiprows=1, delimiter=',',usecols=[1,2,3,4,5])
iris = iris_data[:,:4]
labels = iris_data[:,4]

#sklearn的PCA实现
pca = PCA(n_components=2)
pca_x2d_sklearn = pca.fit_transform(iris)
# print(pca.explained_variance_ratio_)

#手算实现
iris_x = iris - iris.mean(axis=0)
iris_cov = np.cov(iris_x, rowvar=False)
eig_vals, eig_vecs = np.linalg.eig(iris_cov)
index = np.argsort(-np.abs(eig_vals))
pca_x2d_hand = iris_x.dot(eig_vecs[:,index[:2]])

#SVD分解实现
U,D,VT = np.linalg.svd(iris_x)
W = VT.T[:,:2]
pca_x2d_svd = iris.dot(W)


print('sklearn'+'.'*30)
print(pca_x2d_sklearn[:5,:])
print('hand'+'.'*30)
print(pca_x2d_hand[:5,:])
print('SVD'+'.'*30)
print(pca_x2d_svd[:5,:])

print