HNSW is a data structure for very fast, approximate nearest-neighbor searches, especially useful when numeric pairwise distance calculations must scale to large datasets. section 10.2.5
A typical workflow creates an index with a distance function, inserts records individually, and queries the index for a chosen number of neighbors. Approximation improves speed, but results should be interpreted within the needs of the downstream detector. section 10.2.5
Listing 10.7 Example using HNSW for nearest neighbors queries
from datasketch.hnsw import HNSW
import numpy as np
data = np.random.random_sample((1000, 10)) #1
index = HNSW(distance_func=lambda x, y: np.linalg.norm(x - y))
for i, d in enumerate(data): #2
index.insert(i, d)
index.query(data[0], k=10) #3
Shows a concise Python implementation using random 1,000-row, 10-feature numeric data, NumPy distance calculation, insertion, and a ten-neighbor query. section 10.2.5
