Binary vectors - querying the nearest neighbors by Hamming distance with HNSW gives wrong results

TL;DR: [ankane (maintainer)]: Thanks, it looks like the index is working fine. Approximate indexing won't provide perfect recall. Also, you may see better recall by splitting unique hashes into a separate table, as well as increasing

[ankane (maintainer)]: Thanks, it looks like the index is working fine. Approximate indexing won't provide perfect recall. Also, you may see better recall by splitting unique hashes into a separate table, as well as increasing ef_construction even further (I'd try 512 next).

Context: GitHub issue pgvector/pgvector#972 (closed, 8 comments): Hi. So, I have pgvector 0.8.1, postgres 17.7, and a table like this with ~20kk records: ``ruby create_table :image_hashes do |t| t.column :external_id, :integer, null: false t.column :vector_binary, :bit, limit: 576 t.timestamps end add_index :image_hashes, :external_id, unique: true execute 'CREATE INDEX ON image_hashes USING hnsw (vector_binary bit_hamming_ops);' I clearly see that the needed record is in the table: SELECT (vector_binary [~] '[576 bits omitted]') AS distance FROM image_hashes WHERE external_id = 608810322 LIMIT 1; distance ---------- 0 (1 row) `` But querying the n

When to use

You hit exactly this: Binary vectors - querying the nearest neighbors by Hamming distance with HNSW gives wrong results in Binary.

When not to use

A different error, or the same symptom in a different tool. This page only covers the failure above.

Compatibility

Binary (versions mentioned here: 0.8.1, 17.7).

Edge cases

• Also, you may see better recall by splitting unique hashes into a separate table, as well as increasing.

GitHub issue pgvector/pgvector#972 (closed, 8 comments): Hi. So, I have pgvector 0.8.1, postgres 17.7, and a table like this with ~20kk records: ``ruby create_table :image_hashes do |t| t.column :external_id, :integer, null: false t.column :vector_binary, :bit, limit: 576 t.timestamps end add_index :image_hashes, :external_id, unique: true execute 'CREATE INDEX ON image_hashes USING hnsw (vector_binary bit_hamming_ops);' I clearly see that the needed record is in the table: SELECT (vector_binary <~> '[576 bits omitted]') AS distance FROM image_hashes WHERE external_id = 608810322 LIMIT 1; distance ---------- 0 (1 row) But querying the n Verified answer: [ankane (maintainer)]: Thanks, it looks like the index is working fine. Approximate indexing won't provide perfect recall. Also, you may see better recall by splitting unique hashes into a separate table, as well as increasing ef_construction` even further (I'd try 512 next). Source: https://github.com/pgvector/pgvector/issues/972