diskannpy-rust 0.49.1

Python bindings for DiskANN approximate nearest neighbor search library

DiskAnnPy

Python bindings for DiskANN, providing efficient vector quantization and approximate nearest neighbor search.

Installation

Prerequisites

• Python 3.9+
• Rust toolchain (install via rustup)
• maturin (pip install maturin)

Building from Source

Navigate to diskann/DiskAnnPy:

Linux

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate

# Install build dependencies
pip install maturin numpy

# Option 1: Build wheel (creates .whl in DiskANN/target/wheels)
maturin build -r

# Option 2: Build and install directly into active environment
maturin develop -r

Windows

# Create and activate virtual environment
python -m venv venv
venv\Scripts\activate

# Install build dependencies
pip install maturin numpy

# Option 1: Build wheel (creates .whl in DiskANN\target\wheels)
maturin build -r

# Option 2: Build and install directly into active environment
maturin develop -r

Running Tests

Once installed, run the Python tests from the DiskAnnPy directory:

python -m unittest discover tests

Quantizers

See src/quantization/README.md for documentation on the vector quantization APIs.