余弦相似度判断

1.两个文件相似度比较

文本文件

1）先切词，编码把文本向量化处理

2）使用余弦定理计算

二进制文件

读取文件数值，直接判断

#!/usr/bin/python3
# -*- coding: UTF-8 -*-
import sys
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity

with open(sys.argv[1], 'rb') as f1:
pcm1 = np.array([int(_) for _ in f1.read()])

with open(sys.argv[2], 'rb') as f2:
pcm2 = np.array([int(_) for _ in f2.read()])

if np.size(pcm1) > np.size(pcm2):
    delta = np.size(pcm1) - np.size(pcm2)
    pcm2 = np.pad(pcm2, ((0, delta)), mode='constant')
elif np.size(pcm1) < np.size(pcm2):
    delta = np.size(pcm2) - np.size(pcm1)
    pcm1 = np.pad(pcm1, ((0, delta)), mode='constant')
else:
    delta = np.array(0)

sim = cosine_similarity(np.expand_dims(pcm1, axis=0), np.expand_dims(pcm2, axis=0))
print('两个pcm相似度：', sim[0][0])