Add a tutorial for IndexHNSW

[JayjeetAtGithub]

Summary

There are tutorials present for Flat, and IVF but not for HNSW. I believe having a tutorial on HNSW would be helpful.

[mdouze]

right, adding this as enhancement.

[hammad7]

@JayjeetAtGithub

import numpy as np
import faiss

# Define the dimension of the vectors
d = 64

# Define the size of the database
nb = 100000

# Define the number of queries
nq = 10000

# Set the random seed for reproducibility
np.random.seed(1234)

# Generate random vectors for the database
xb = np.random.random((nb, d)).astype('float32')

# Add a unique identifier to each vector in the database
xb[:, 0] += np.arange(nb) / 1000.

# Generate random vectors for the queries
xq = np.random.random((nq, d)).astype('float32')

# Add a unique identifier to each vector in the queries
xq[:, 0] += np.arange(nq) / 1000.

# Create an HNSW index
index = faiss.IndexHNSWFlat(d, 32)

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

# Perform a search
D, I = index.search(xq, 4)

# Print the results
print(I[:5])  # Neighbors of the first 5 queries
print(I[-5:])  # Neighbors of the last 5 queries

[mlomeli1]

Re-opening since the python tutorial is pending.