MRPT
MRPT (Multi-Resolution Proximity Trees) is an open-source library for fast approximate nearest neighbor search in high-dimensional vector spaces, applicable to vector database backends.
About this tool
MRPT
MRPT (Multi-Resolution Proximity Trees) is an open-source library for fast approximate nearest neighbor (ANN) search in high-dimensional vector spaces. It is written in C++11 and provides Python bindings. MRPT is designed to be lightweight, easy to use, and efficient, making it suitable for vector database backends and other applications requiring ANN search.
- Source: GitHub Repository
- Category: Open-source
- Tags: open-source, ann, high-dimensional, vector-search
Features
- Approximate Nearest Neighbor Search: Efficient ANN search in high-dimensional spaces.
- C++11 Implementation: Core library written in modern C++ (header-only).
- Python Bindings: Easy integration with Python via pip package (
pip install mrpt). - Random Projection Trees: Uses a collection of random projection trees to index data for fast querying.
- Automatic Hyperparameter Tuning: Integrated autotuning algorithm; only the target recall level and number of neighbors need to be specified. No manual hyperparameter tuning required.
- Euclidean Distance Metric: Currently supports Euclidean distance for similarity measurement.
- Minimal Dependencies: Only dependency is Eigen (linear algebra library), bundled with the library.
- No Compilation Required (Header-only): For C++, just include the header file.
- Examples Provided: Minimal usage examples for both Python and C++.
- Open Source: MIT license.
Usage
- Python: Install via
pip install mrpt. - C++: Include the
cpp/Mrpt.hheader and link with Eigen.
Documentation
- C++ API documentation
- Algorithm description and complexity analysis
- Automatic hyperparameter tuning article
Pricing
- Free and open-source (MIT License)
Loading more......
Information
Categories
Similar Products
6 result(s)An open-source library for approximate nearest neighbor search in high-dimensional spaces, often used as a backend for vector databases and search engines.
KGraph is an open-source library for fast approximate nearest neighbor search in high-dimensional vector spaces, applicable to vector database solutions.
NGT (Neighborhood Graph and Tree) is an open-source vector search engine designed for fast and scalable approximate nearest neighbor search.
Voyager is a Spotify open-source vector search library and service for efficient nearest neighbor search on large-scale vector datasets.
vsag is an Alibaba open-source library implementing efficient vector search algorithms, including approximate nearest neighbor search for high-dimensional vectors.
Cottontail DB is an open-source vector database for storing and searching high-dimensional data, with features geared towards research and production environments.