E-commerce Product Recommendations FAISS

import faiss
import numpy as np

# Assume we have product embeddings (vectors) for all our products
product_embeddings = np.random.random((1000000, 128)).astype('float32')  # 1M products, 128-dimensional embeddings

# Create an index
index = faiss.IndexFlatL2(128)  # Using L2 distance

# Add vectors to the index
index.add(product_embeddings)

# Let's say a user is viewing a product with this embedding
query_embedding = np.random.random((1, 128)).astype('float32')

# Search for similar products
k = 5  # Number of similar products to retrieve
distances, indices = index.search(query_embedding, k)

print(f"Indices of similar products: {indices}")
print(f"Distances to similar products: {distances}")