import numpy as np [2]: import matplotlib.pyplot as plt [3]: np.random.seed(42) [6]: X=np.random.randn(10) [7]: X [7]: array([-0.56228753, -1.01283112, 0.31424733, -0.90802408, -1.4123037 , 1.46564877, -0.2257763 , 0.0675282 , -1.42474819, -0.54438272]) [8]: Y=X+np.random.randn(10) [9]: Y [9]: array([-0.45136494, -2.1638247 , 0.68994535, -1.50866277, -1.70399745, 0.86394216, 1.62650188, 0.05403098, -2.48245912, 0.27816219]) [18]: plt.scatter(X,Y) plt.xlabel('X Values') plt.ylabel('Y Values') plt.show() 1 0.0.1 Pearsons Correlation [19]: from scipy.stats import pearsonr [20]: corr,_=pearsonr(X,Y) [21]: print('Pearsons correlation: %.3f' %corr) Pearsons correlation: 0.778 0.0.2 Cosine Similarity [22]: from sklearn.metrics.pairwise import cosine_similarity [23]: cos_sim = cosine_similarity(X.reshape(1,-1),Y.reshape(1,-1)) [24]: print('Cosine Similarity: %.3f' % cos_sim) Cosine Similarity: 0.805 2 0.0.3 Jaccard Similarity [25]: from sklearn.metrics import jaccard_score [26]: A = [1,1,1,0] [28]: B = [1,1,0,0] [29]: jacc = jaccard_score(A,B) [30]: print('Jaccard Similarity: %.3f' %jacc) Jaccard Similarity: 0.667 0.0.4 Euclidean Distance [31]: from scipy.spatial import distance [32]: dst = distance.euclidean(X,Y) [35]: print('Euclidean distance: %.3f' % dst) Euclidean distance: 2.741 0.0.5 Manhattan Distance [36]: from scipy.spatial import distance [37]: dst = distance.cityblock(X,Y) [38]: print('Manhattan distance: %.3f' % dst) Manhattan distance: 6.878