vectorized Python implementation using NumPy to compute the cosine similarity between every pair of vectors in a list

cosin.py

import numpy as np

def cosine_similarity_matrix(vectors):
    """
    Compute the cosine similarity matrix between all pairs of vectors.
    
    Parameters:
        vectors (List or np.ndarray): A 2D array of shape (n_vectors, vector_dim).
    
    Returns:
        np.ndarray: A 2D array of shape (n_vectors, n_vectors) with cosine similarity scores.
    """
    vectors = np.array(vectors)
    
    # Normalize the vectors (avoid division by zero)
    norms = np.linalg.norm(vectors, axis=1, keepdims=True)
    norms[norms == 0] = 1e-10  # Handle zero vectors
    
    normalized = vectors / norms
    
    # Cosine similarity = dot product of normalized vectors
    similarity_matrix = np.dot(normalized, normalized.T)
    
    return similarity_matrix

# Example usage:
vectors = [
    [1, 0, 0],
    [0, 1, 0],
    [1, 1, 0],
]

cosine_scores = cosine_similarity_matrix(vectors)
print(cosine_scores)