The spatial.cosine.distance() function from the scipy module calculates the distance instead of the cosine similarity, but to achieve that, we can subtract the value of the distance from 1.
from scipy import spatial
List1 = [4, 47, 8, 3]
List2 = [3, 52, 12, 16]
result = 1 - spatial.distance.cosine(List1, List2)
print(result)
The numpy.dot() function calculates the dot product of the two vectors passed as parameters. The numpy.norm() function returns the vector norm.
I can use these functions with the correct formula to calculate the cosine similarity.
from numpy import dot
from numpy.linalg import norm
List1 = [4, 47, 8, 3]
List2 = [3, 52, 12, 16]
result = dot(List1, List2)/(norm(List1)*norm(List2))
print(result)
In the sklearn module, there is an in-built function called cosine_similarity() to calculate the cosine similarity.
from sklearn.metrics.pairwise import cosine_similarity,cosine_distances
A=np.array([10,3])
B=np.array([8,7])
result=cosine_similarity(A.reshape(1,-1),B.reshape(1,-1))
print(result)
When I deal with N-dimensional tensors having shapes (m,n), I can use the consine_similarity() function from the torch module to find the cosine similarity.
import torch
import torch.nn.functional as F
t1 = [3,45,6,8]
a = torch.FloatTensor(t1)
t2 = [4,54,3,7]
b = torch.FloatTensor(t2)
result = F.cosine_similarity(a, b, dim=0)
print(result)
Which method should I use to get the best similarity results if I use 2d array using the above modules in python?
