annoy VS hnswlib

The focus on the top 10 in vector search is a product of wanting to prove value over keyword search. Keyword search is going to miss some conceptual matches. You can try to work around that with tokenization and complex queries with all variations but it's not easy.

Vector search isn't all that new a concept. For example, the annoy library (https://github.com/spotify/annoy) has been around since 2014. It was one of the first open source approximate nearest neighbor libraries. Recommendations have always been a good use case for vector similarity.

Recommendations are a natural extension of search and transformers models made building the vectors for natural language possible. To prove the worth of vector search over keyword search, the focus was always on showing how the top N matches include results not possible with keyword search.

In 2023, there has been a shift towards acknowledging keyword search also has value and that a combination of vector + keyword search (aka hybrid search) operates in the sweet spot. Once again this is validated through the same benchmarks which focus on the top 10.

On top of all this, there is also the reality that the vector database space is very crowded and some want to use their performance benchmarks for marketing.

Disclaimer: I am the author of txtai (https://github.com/neuml/txtai), an open source embeddings database

If you want to go larger you could still use some simple setup in conjunction with faiss, annoy or hnsw.

I then use annoy to compare them. Annoy can use different measures for distance, like cosine, euclidean and more

Yes you can do this for equality predicates if your row groups are sorted . This blog post (that I didn't write) might add more color. You can't do this for any kind of text searching. If you need to do this with file based storage I'd recommend using a vector based text search and utilize a ANN index library like Annoy.

If you need large scale (1000+ dimension, millions+ source points, >1000 queries per second) and accept imperfect results / approximate nearest neighbors, then other people have already mentioned some of the best libraries (FAISS, Annoy).

Would be possible to further speed up the process with using something like ANNOY? https://github.com/spotify/annoy

I like Faiss but I tried Spotify's annoy[1] for a recent project and was pretty impressed.

Since lots of people don't seem to understand how useful these embedding libraries are here's an example. I built a thing that indexes bouldering and climbing competition videos, then builds an embedding of the climber's body position per frame. I then can automatically match different climbers on the same problem.

It works pretty well. Since the body positions are 3D it works reasonably well across camera angles.

The biggest problem is getting the embedding right. I simplified it a lot above because I actually need to embed the problem shape itself because otherwise it matches too well: you get frames of people in identical positions but on different problems!

[1] https://github.com/spotify/annoy

If you just want quick in memory search then pynndescent is a decent option: it's easy to install, and easy to get running. Another good option is Annoy; it's just as easy to install and get running with python, but it is a little less performant if you want to do a lot of queries, or get a knn-graph quickly.

How does this compare to hsnwlib - is it faster? https://github.com/nmslib/hnswlib

Last week I released Moodflix (https://moodflix.streamlit.app), a movie recommendation engine based to find movies based on your mood.

Moodflix was created on top of a movie dataset of 10k movies from The Movie Database. I vectorised the films using Hugging Face's T5 model (https://huggingface.co/docs/transformers/model_doc/t5) using the film's plot synopsis, genres and languages. Then I indexed the vectors using hnswlib (https://github.com/nmslib/hnswlib). LLMs can understand a movie's plot pretty well and distill the similarities between a user's query (mood) to the movie's plot and genres.

I have got feedback from close friends around linking movies to other review sites like IMDB or Rotten Tomatoes, linking movies to sites to stream the movie and adding movie posters. I would also love to hear from the community what things you like, what you want to see and what things you consider can be improved.

Actually the "ef" is not epsilon. It is a parameter of the HNSW index: https://github.com/nmslib/hnswlib/blob/master/ALGO_PARAMS.md...

If you want to go larger you could still use some simple setup in conjunction with faiss, annoy or hnsw.

Many vector databases are using Hnswlib and that is a supported vector index alongside Faiss and Annoy.

hnswlib (https://github.com/nmslib/hnswlib) is a strong alternative to faiss that I have enjoyed using for multiple projects. It is simple and has great performance on CPU.

After working through several projects that utilized local hnswlib and different databases for text and vector persistence, I integrated hnswlib with sqlite to create an embedded vector search engine that can easily scale up to millions of embeddings. For self-hosted situations of under 10M embeddings and less than insane throughput I think this combo is hard to beat.

https://github.com/jiggy-ai/hnsqlite

https://github.com/nmslib/hnswlib

Used it to index 40M text snippets in the legal domain. Allows incremental adding.

I love how it just works. You know, doesn’t ANNOY me or makes a FAISS. ;-)

3000 texts doesn't sound like to many, so may be a brute force cos calculation to find the most similar vector would work. If that's taking too much time, may be look at KNN or ANN modules to speed up finding the most similar vector. I use hsnwlib in knn mode for this. SOrt through about 350,000 vectors in about 30-50 msec.

hnswlib is in cpp and has python bindings (you should be able to make your own for other languages).

https://github.com/nmslib/hnswlib

embeddings - The embeddings index file. This is an Approximate Nearest Neighbor (ANN) index with either Faiss (default), Hnswlib or Annoy, depending on the settings.