autofaiss
autofaiss copied to clipboard
criteo
Automatically create Faiss knn indices with the most optimal similarity search parameters.
Bumps [mypy](https://github.com/python/mypy) from 1.8.0 to 1.9.0. Changelog Sourced from mypy's changelog. Mypy Release Notes Mypy 1.9 We’ve just uploaded mypy 1.9 to the Python Package Index (PyPI). Mypy is a...
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot] avatar"){kind=avatar}
Bumps [pylint](https://github.com/pylint-dev/pylint) from 3.0.3 to 3.1.0. Commits 053c2c3 Bump pylint to 3.1.0, update changelog c954636 Upgrade release documentation, and contributors.txt 7300ed2 Discover .pyi files (#9241) 9dbf3df Merge maintenance 3.0.x into...
While adding this content to the tarball is correct and appropriate behavior, using `data_files` for the purpose means that these files are installed _into the `site-packages` directory that autofaiss is...
Some IVF KNN indices built by autofaiss have bad recall scores due to the unreachable speed constraints given in the requirements. A way to guarantee decent recall scores is to...
Hello everyone! Thank you for this nice project and the features already developed. I am currently trying to create a big faiss index in a distributed way and found that...
Hi, I had very bad scores using the autofaiss `build_index` method and except that the bug could have been caused from the fact that when you `build_index` you don't save...
We are using autofaiss to build a pipeline for creating indexes of different sizes. We have embedding datasets ranging from 1M to 1B examples. Currently all the embeddings we use...
Reading [these docs](https://criteo.github.io/autofaiss/getting_started/quantization.html) it appears as though one can set the entry IDs when using parquet by setting `–id_columns`. How does one set entry IDs when using npy format?
`autofaiss` [currently requires](https://github.com/criteo/autofaiss/blob/1a7b55bad90d56e33f8c6b8fa222e3226429c461/requirements.txt#L4) that `pandas` be installed at a version below 2.0. This is problematic, because many new libraries use pandas>=2.0. Is it possible to change the requirements file to...
