The script calculates covariance matrix, eigenvectors and eigenvalues

pca_test.py

#!/usr/bin/env python3

import numpy as np
import pandas as pd

## Import Wine dataset
df_wine = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/wine/wine.data', header=None)

# data.iloc[<row selection>, <column selection>] the features are extracted and the labels are ignored (column 0)
X = df_wine.iloc[:, 1:].values
print("Number of dataset feautures: ", len(X.T)) # should be 13

## Covariance matrix, eigenvectors and eigenvalues
C = np.cov(X.T)
eigenvalues, eigenvectors = np.linalg.eig(C)

## Dataset Transformation
eigenvalues2eigenvectors = [(eigenvalues[i], eigenvectors[:,i]) for i in range(len(eigenvalues))]
eigenvalues2eigenvectors.sort(reverse=True)

W = np.hstack( (eigenvalues2eigenvectors[0][1][:, np.newaxis], eigenvalues2eigenvectors[1][1][:, np.newaxis]))
X_PCA = X.dot(W)
print("Number of dataset feautures: ", len(X_PCA.T)) # should be 2