ann-benchmarks
ann-benchmarks copied to clipboard
erikbern
Benchmarks of approximate nearest neighbor libraries in Python
hi， I use the ann-benchmarks parameters(R=64mL=200,level=1) and then could not get the same result performance, BTW，I use latest glassppy for python version, the best recall is about 90%.
We've updated the kgn test scripts. Could you please review and merge them when you have a moment? Thanks for your help!
## TL;DR - 🏆 **99.0% Recall@10** + **27,857 QPS** achieved - 📊 **Beat industry standards** by 10-40% across all metrics - 🔒 **IP protected** with Docker blackbox (no source code...
# Add Quark Platform algorithms for ann-benchmarks ## 🎯 Overview This PR adds Quark Platform algorithms to ann-benchmarks with complete IP protection using Docker blackbox approach. ## ✅ Algorithms Added...
This pull request adds support for the [PASE](PostgreSQL Approximate Search Extension) algorithm to the ANN-Benchmarks framework. **What's included:** - New algorithm folder: `ann_benchmarks/algorithms/pase/` - `module.py`: Python wrapper for PASE -...
Downgrade marshmallow to avoid import errors
Hello everyone, I am running ann-benchmarks using pgvector algorithm/extension on different datasets provided by this repository and with different parameters for the hnsw index. At first, I have tested it...
Hi, Erik Here are several issues are seen when I run algorithm with --local and --parallelism > 1, could you please take a look? 1. Milvus is failed to create...
I upgraded the pybind11 version used in NGT. The previous version was outdated, causing an issue where all elements in the search result array had the same value. Additionally, for...
