Python及linux部分使用技巧

1、計算詞向量矩陣彼此間餘弦相似度
即由nm的詞向量矩陣得到nn的相似度矩陣

import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
vec1=[0.1,0.5,0.8]
vec2=[0.1,0.2,0.33]
cos2=cosine_similarity(x)

現象

2、Python使用networkx庫計算拉普拉斯矩陣
其中similarity_matrix即爲輸入的權值方陣

	similarity_matrix = np.load(load_url)
    m=np.matrix(similarity_matrix)
    G = nx.from_numpy_matrix(m)
    print(nx.laplacian_matrix(G).toarray())

3、生成矩陣對應的圖像

from PIL import Image  
matrix = matrix*255
    matrix = np.matrix(matrix)
    one = np.ones((574,574))*120
    matrix=matrix+one
    print(matrix)
    matrix_image = Image.fromarray(matrix)
    matrix_image=matrix_image.resize((500,800))
    matrix_image=matrix_image.convert("L")#轉換成灰度圖
    matrix_image.save('image.png')

4、