simonw / simonw/llm

Improvement of cosine similarity

Open
#233 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
12.5k
Forks
991
Avg merge
3d 13h
Merged PRs (30d)
10

Description

I saw that currently the cosine similarity is computed on a vector by vector basis, which can become quite slow for a big amount of samples.
I recently had the same issue on a personal project with around 9k embeddings; I ended up using sklearn's pairwise cosine similarity, and that was much faster as it computes the whole matrix at once.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Locate the current vector-by-vector cosine similarity implementation and inspect how embeddings are passed through it. Compare its results and performance with scikit-learn's pairwise cosine similarity on a large embedding set; done means preserving expected similarity values while improving performance for many samples.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, scikit-learn
Domain
machine-learning
Issue type
Refactor
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.