主成分分析-python机器学习实现(PCA)

代码思路来自Python Machine Learning 3rd，本人只是做了一些修改

numpy + pandas实现

我们以pandas内置的wine库来实现，我这里采取的是本地读取

import pandas as pd
df_wine = pd.read_csv('./wine.data')

1. 标准化原始数据集

# 首先分离数据集
from sklearn.model_selection import train_test_split
X, y = df_wine.iloc[:, 1:], df_wine.iloc[:, 0]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, stratify=y, random_state=0)

# 标准化
from sklearn.preprocesssing import StandardScalor
sc = StandardScalor()
X_train_std = sc.fit_transform(X_train)
X_test_std = sc.transform(X_test)

2. 获得协方差矩阵

使用numpy API

import numpy as np
cov_mat = np.cov(X_train_std.T)

3. 获得特征向量和特征值

使用numpy API

eigen_vals, eigen_vecs = np.linalg.eigh(