同样都是欧式距离的平方，结果会不一样？

东方耀

同样都是欧式距离的平方，结果会不一样？






import numpy as np
# 为何这两种计算会不相等


embeddings1 = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
embeddings2 = np.array([[2, 3, 5], [1, 4, 3], [6, 10, 9]])


print(embeddings1.shape, embeddings2.shape)


diff = np.subtract(embeddings1, embeddings2)
dist = np.sum(np.square(diff), 1)
print(dist)


dist_dfy = np.square(np.linalg.norm(diff, axis=1))
print(dist_dfy)





2个向量是否完全相等(欧式距离的平方)因为后者是先开了根号后平方的：会有精度的丢失

东方耀

直接计算  平方和  速度还要快些