Lightweight vector database.
Project description
VectorLiteDB
VectorLite is a simple vector database powered by sentence-transformers
and hnswlib
, designed for efficient storage and search of textual data.
🚀 Quick Install
pip install vectorlite
Usage
Run as api server
-
serve vectorlite
-
Navigate to http://localhost:4440/docs
Run integrated in code
Initialization
from vectorlite import VectorLite
vl = VectorLite()
Adding Data
data = ["Sample text 1", "Sample text 2", "Another example"]
vl.create(data)
Read All Data
all_data = vl.read_all()
To limit the number of records returned:
limited_data = vl.read_all(max_items=2)
Read Specific Data by Index
item = vl.read(1)
Update Data by Index
vl.update(1, "Updated sample text 2")
Delete Data by Index
vl.delete(1)
Similarity Search
results = vl.similarity_search("A related sample text", k=3)
Semantic Search
results = vl.semantic_search("A related sample text", k=3)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
vectorlite-0.2.9.tar.gz
(7.8 kB
view hashes)
Built Distribution
Close
Hashes for vectorlite-0.2.9-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | f2b6010c6f5ebc2f177ee4c73f9340c00571d630206a9a5bcbdc6b077fc45423 |
|
MD5 | 897aec0e1df102f3ac9b6228228c0ab4 |
|
BLAKE2b-256 | f146b0861c2125b0ed3fbca87a3281471039deb818d3878038975d9679b7fc8c |