Python：PCA代码实现

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

class PrincipalComponentAnalysis(object):
    def __init__(self,X=None, k=None):
        self.X = X
        self.k = k

    def fit(self,X,k):
        self.X = X
        self.k = k
        self.nSample, self.nDim = self.X.shape
        Xmean = np.mean(self.X,axis=0)
        self.Xnorm = self.X - Xmean

        scatter_matric = self.Xnorm.T @ self.Xnorm
        eig_val, eig_vec = np.linalg.eig(scatter_matric)
        ord_ids = np.flipud(np.argsort(eig_val))
        feature = np.zeros((self.nDim,self.k))
        for i in range(self.k):
            feature[:,i] = eig_vec[:,ord_ids[i]]

        data = self.Xnorm @ feature
        return data

if __name__ == '__main__':
    X,y = datasets.load_iris(return_X_y=True)
    pca = PrincipalComponentAnalysis()
    data = pca.fit(X=X,k=2)
    print(data.shape)



    plt.rcParams['font.sans-serif'] = ['SimHei']  # 步骤一（替换sans-serif字体）
    plt.rcParams['axes.unicode_minus'] = False  # 步骤二（解决坐标轴负数的负号显示问题）


    fig = plt.figure(figsize=(10,5))
    ax1 = fig.add_subplot(121)
    ax2 = fig.add_subplot(122)
    ax1.scatter(data[:,0],data[:,1],c=y)
    ax1.set_title("python实现")
    ax1.set_xlabel("x轴")
    ax1.set_ylabel("y轴")


    pca1 = PCA(n_components=2)
    data1 = pca1.fit_transform(X=X)
    ax2.scatter(data1[:,0],-data1[:,1],c=y)
    ax2.set_title("调包实现")
    plt.show()

参考：

https://zhuanlan.zhihu.com/p/47858230

https://zhuanlan.zhihu.com/p/77151308