Skip to main content

Product quantization for nearest neighbor search in a single python file

Project description

Copyright (c) 2018 Yusuke Matsui

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Description: # nanopq

[![Documentation Status](https://readthedocs.org/projects/nanopq/badge/?version=latest)](https://nanopq.readthedocs.io/en/latest/?badge=latest)

Nano Product Quantization (nanopq): product quantization for nearest neighbor search in a single python file.

This package contains a vanilla implementation of Product Quantization (PQ) and Optimized Product Quantization (OPQ) written in pure python without any third party dependencies.


## Installing
You can install the package via pip. This library works with Python 3.6+ on linux.
```
pip install nanopq
```

## [Documentation](https://nanopq.readthedocs.io/en/latest/index.html)
- [Tutorial](https://nanopq.readthedocs.io/en/latest/tutorial.html)
- [API](https://nanopq.readthedocs.io/en/latest/api.html)

## Example

```python
import nanopq
import numpy as np

N, D = 10000, 128
X = np.random.random((N, D)).astype(np.float32) # 10,000 128-dim vectors
query = np.random.random((D,)).astype(np.float32) # a 128-dim vector

# Instantiate with M=8 sub-spaces
pq = nanopq.PQ(M=8)

# Train with the top 1000 vectors
pq.fit(X[:1000])

# Encode to pq-codes
X_code = pq.encode(X) # (10000, 8) with dtype=np.uint8

# Results
dtable = pq.dtable(query) # Compute a distance table online
dists = pq.adist(dtable, X_code) # Asymmetric distance
```

## Author
- [Yusuke Matsui](http://yusukematsui.me)


## Reference
- [H. Jegou, M. Douze, and C. Schmid, "Product Quantization for Nearest Neighbor Search", IEEE TPAMI 2011](https://ieeexplore.ieee.org/document/5432202/) (the original paper of PQ)
- [T. Ge, K. He, Q. Ke, and J. Sun, "Optimized Product Quantization", IEEE TPAMI 2014](https://ieeexplore.ieee.org/document/6678503/) (the original paper of OPQ)
- [Y. Matsui, Y. Uchida, H. Jegou, and S. Satoh, "A Survey of Product Quantization", ITE MTA 2018](https://www.jstage.jst.go.jp/article/mta/6/1/6_2/_pdf/) (a survey paper of PQ)
- [PQ in faiss](https://github.com/facebookresearch/faiss/wiki/Faiss-building-blocks:-clustering,-PCA,-quantization#pq-encoding--decoding) (Faiss contains an optimized implementation of PQ. [See the difference to ours here](https://nanopq.readthedocs.io/en/latest/tutorial.html#difference-from-pq-in-faiss))
- [Rayuela.jl](https://github.com/una-dinosauria/Rayuela.jl) (Julia implementation of several encoding algorithms including PQ and OPQ)
- [PQk-means](https://github.com/DwangoMediaVillage/pqkmeans) (clustering on PQ-codes. The implementation of nanopq is compatible to [that of PQk-means](https://github.com/DwangoMediaVillage/pqkmeans/blob/master/tutorial/1_pqkmeans.ipynb))
Platform: UNKNOWN
Description-Content-Type: text/markdown

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nanopq-0.1.2.tar.gz (6.2 kB view details)

Uploaded Source

Built Distribution

nanopq-0.1.2-py3-none-any.whl (7.2 kB view details)

Uploaded Python 3

File details

Details for the file nanopq-0.1.2.tar.gz.

File metadata

  • Download URL: nanopq-0.1.2.tar.gz
  • Upload date:
  • Size: 6.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for nanopq-0.1.2.tar.gz
Algorithm Hash digest
SHA256 eb33c1c069c70304ef02955b9fb3b32ac974926b77e7e99ac87d215b49276c95
MD5 9cd2484a49b749245da057be42189010
BLAKE2b-256 cfaab52c01bb42253a22df4ea218d05d38de27ba6b6091a42d8f27959c040d78

See more details on using hashes here.

File details

Details for the file nanopq-0.1.2-py3-none-any.whl.

File metadata

File hashes

Hashes for nanopq-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 074505ac868b42f8f0b9a0aad31f2cb294076c35b4c280ef2b63fd7d2371d578
MD5 fcdefee45813610145d66e1560f638dc
BLAKE2b-256 18978ca7257f0574252db5cad618650215dbf7d5d73807040b422a50f69314db

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page