https://github.com/hora-search/horapy
🐍 Python bidding for the Hora Approximate Nearest Neighbor Search Algorithm library
https://github.com/hora-search/horapy
approximate-nearest-neighbor-search artificial-intelligence image-search python python-library recommendation-system search search-engine
Last synced: 7 months ago
JSON representation
🐍 Python bidding for the Hora Approximate Nearest Neighbor Search Algorithm library
- Host: GitHub
- URL: https://github.com/hora-search/horapy
- Owner: hora-search
- License: apache-2.0
- Created: 2021-06-13T21:36:15.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-10-11T00:53:15.000Z (about 4 years ago)
- Last Synced: 2024-08-01T16:47:40.586Z (about 1 year ago)
- Topics: approximate-nearest-neighbor-search, artificial-intelligence, image-search, python, python-library, recommendation-system, search, search-engine
- Language: Python
- Homepage: https://horasearch.com
- Size: 95.7 KB
- Stars: 69
- Watchers: 5
- Forks: 10
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# horapy
**[[Homepage](http://horasearch.com/)]** **[[Document](https://horasearch.com/doc)]** **[[Examples](https://horasearch.com/doc/example.html)]** **[[Hora](https://github.com/hora-search/hora)]**
Python binding for the **`Hora Approximate Nearest Neighbor Search`**
# Key Features
* **Performant** ⚡️
* **SIMD-Accelerated ([packed_simd](https://github.com/rust-lang/packed_simd))**
* **Stable algorithm implementation**
* **Multiple threads design**
* **Multiple Indexes Support** 🚀
* `Hierarchical Navigable Small World Graph Index(HNSWIndex)` ([detail](https://arxiv.org/abs/1603.09320))
* `Satellite System Graph (SSGIndex)` ([detail](https://arxiv.org/abs/1907.06146))
* `Product Quantization Inverted File(PQIVFIndex)` ([detail](https://lear.inrialpes.fr/pubs/2011/JDS11/jegou_searching_with_quantization.pdf))
* `Random Projection Tree(RPTIndex)` (LSH, WIP)
* `BruteForce (BruteForceIndex)` (naive implementation with SIMD)
* **Portable** 💼
* Support `no_std` (WIP, partial)
* Support `Windows`, `Linux` and `OS X`
* Support `IOS` and `Android` (WIP)
* **No** heavy dependency, such as `BLAS`
* **Reliability** 🔒
* `Rust` compiler secure all code
* Memory managed by `Rust`
* Broad testing coverage
* **Multiple Distances Support** 🧮
* `Dot Product Distance`
* 
* `Euclidean Distance`
* 
* `Manhattan Distance`
* 
* `Cosine Similarity`
* 
* **Productive** ⭐
* Well documented
* Elegant and simple API, easy to learn
# Benchmark
by `aws t2.medium (CPU: Intel(R) Xeon(R) CPU E5-2686 v4 @ 2.30GHz)` [more information](https://github.com/hora-search/ann-benchmarks)
## Installation
```bash
pip install horapy
```
## Example
```Python
import numpy as np
from horapy import HNSWIndex
dimension = 50
n = 1000
# init index instance
index = HNSWIndex(dimension, "usize")
samples = np.float32(np.random.rand(n, dimension))
for i in range(0, len(samples)):
# add node
index.add(np.float32(samples[i]), i)
index.build("euclidean") # build index
target = np.random.randint(0, n)
# 410 in Hora ANNIndex (dimension: 50, dtype: usize, max_item: 1000000, n_neigh: 32, n_neigh0: 64, ef_build: 20, ef_search: 500, has_deletion: False)
# has neighbors: [410, 736, 65, 36, 631, 83, 111, 254, 990, 161]
print("{} in {} \nhas neighbors: {}".format(
target, index, index.search(samples[target], 10))) # search
```
# License
The entire repo is under [Apache License](https://github.com/hora-search/hora/blob/main/LICENSE).