Metadata-Version: 2.4
Name: patann
Version: 0.0.85
Summary: PatANN is a massively parallel, distributed, and scalable in-memory/on-disk vector database library for efficient nearest neighbor search across large-scale datasets by finding vector patterns.
Home-page: https://github.com/mesibo/patann
Author: https://mesibo.com
Author-email: support@mesibo.com
Project-URL: Bug Tracker, https://github.com/mesibo/patann/issues
Classifier: Development Status :: 5 - Production/Stable
Classifier: Intended Audience :: Developers
Classifier: Operating System :: Unix
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.9
Description-Content-Type: text/markdown
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: project-url
Dynamic: requires-python
Dynamic: summary

# PatANN - Pattern-Aware Vector Database / ANN

## Overview
PatANN is a massively parallel, distributed, and scalable vector database library for efficient nearest neighbor search across large-scale datasets by finding vector patterns.

## Status
**Beta Version**: Currently uploaded for benchmarking purposes. Complete documentation and updates are under development. Not for production use yet.

## Platforms
- **Beta Version**: Restricted to Linux to prevent premature circulation of beta versions
- **Production Releases (late April 2025)***: Will support all platforms that are supported by mesibo

## Key Features
- Faster Index building and Searching
- Supports both in-memory and on-disk operations
- Dynamic sharding to load balance across servers
- Refined search, filtering and pagination
- Unlimited scalability without pre-specified capacity

## Algorithmic Approach
- Novel pattern-based probing technique for ANN search
- Preliminary results show phenomenal performance in building index and searching
- Potential slight variations in lower-end matching
- Detailed research paper forthcoming

## Contributions
We are seeking help to:

- Run additional datasets. So far, all tested datasets (including self-generated) exhibit patterns that helps algorithm. We have yet to test datasets without clear patterns or with uniform distribution.
- Validate and improve the algorithm

## Contact
For support / questions, please contact: support@mesibo.com

