Python client
Manage unstructured vector stores in PostgreSQL.
Supabase provides a Python client called vecs for managing unstructured vector stores. This client provides a set of useful tools for creating and querying collections in PostgreSQL using the pgvector extension.
Quick start#
Let's see how Vecs works using a local database. Make sure you have the Supabase CLI installed on your machine.
Initialize your project#
Start a local Postgres instance in any folder using the init and start commands. Make sure you have Docker running!
_10# Initialize your project_10supabase init_10_10# Start Postgres_10supabase start
Create a collection#
Inside a Python shell, run the following commands to create a new collection called "docs", with 3 dimensions.
_10import vecs_10_10# create vector store client_10vx = vecs.create_client("postgresql://postgres:postgres@localhost:54322/postgres")_10_10# create a collection of vectors with 3 dimensions_10docs = vx.get_or_create_collection(name="docs", dimension=3)
Add embeddings#
Now we can insert some embeddings into our "docs" collection using the upsert() command:
_13import vecs_13_13# create vector store client_13docs = vecs.get_or_create_collection(name="docs", dimension=3)_13_13# a collection of vectors with 3 dimensions_13vectors=[_13 ("vec0", [0.1, 0.2, 0.3], {"year": 1973}),_13 ("vec1", [0.7, 0.8, 0.9], {"year": 2012})_13]_13_13# insert our vectors_13docs.upsert(vectors=vectors)
Query the collection#
You can now query the collection to retrieve a relevant match:
_10import vecs_10_10docs = vecs.get_or_create_collection(name="docs", dimension=3)_10_10# query the collection filtering metadata for "year" = 2012_10docs.query(_10 data=[0.4,0.5,0.6], # required_10 limit=1, # number of records to return_10 filters={"year": {"$eq": 2012}}, # metadata filters_10)
Deep dive#
For a more in-depth guide on vecs collections, see API.
Resources#
- Official Vecs Documentation: https://supabase.github.io/vecs/api
- Source Code: https://github.com/supabase/vecs