pgvector support for Lua
Supports pgmoon
Run:
luarocks install pgvector
And follow the instructions for your database library:
Or check out some examples:
- Embeddings with OpenAI
- Binary embeddings with Cohere
- Hybrid search with Ollama (Reciprocal Rank Fusion)
- Sparse search with Text Embeddings Inference
Require the library
local pgvector = require("pgvector")
Enable the extension
pg:query("CREATE EXTENSION IF NOT EXISTS vector")
Create a table
pg:query("CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3))")
Insert a vector
local embedding = pgvector.new({1, 1, 1}) pg:query("INSERT INTO items (embedding) VALUES ($1)", embedding)
Get the nearest neighbors
local embedding = pgvector.new({1, 1, 1}) local res = pg:query("SELECT * FROM items ORDER BY embedding <-> $1 LIMIT 5", embedding) for i, row in ipairs(res) do print(row["id"]) end
Add an approximate index
pg:query("CREATE INDEX ON items USING hnsw (embedding vector_l2_ops)") -- or pg:query("CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 100)")
Use vector_ip_ops for inner product and vector_cosine_ops for cosine distance
See a full example
Create a half vector from a table
local vec = pgvector.halfvec({1, 2, 3})
Create a sparse vector from a table of non-zero elements
local elements = {[1] = 1, [3] = 2, [5] = 3} local vec = pgvector.sparsevec(elements, 6)
Get the number of dimensions
vec["dim"]
Get the non-zero elements
vec["elements"]
View the changelog
Everyone is encouraged to help improve this project. Here are a few ways you can help:
- Report bugs
- Fix bugs and submit pull requests
- Write, clarify, or fix documentation
- Suggest or add new features
To get started with development:
git clone https://github.com/pgvector/pgvector-lua.git
cd pgvector-lua
createdb pgvector_lua_test
luarocks install pgmoon
luarocks install luasocket
lua test/pgvector.luaTo run an example:
createdb pgvector_example luarocks install luasec luarocks install lua-cjson lua examples/openai/example.lua