philiprehberger-embedding-store 0.1.0

In-memory vector store with cosine similarity search

philiprehberger-embedding-store

In-memory vector store with cosine similarity search.

Install

pip install philiprehberger-embedding-store

Usage

from philiprehberger_embedding_store import VectorStore

store = VectorStore(dimensions=1536)

# Add vectors with metadata
store.add("doc1", embedding=[0.1, 0.2, ...], metadata={"title": "First doc"})
store.add("doc2", embedding=[0.3, 0.1, ...], metadata={"title": "Second doc"})

# Search by similarity
results = store.search(query_embedding=[0.15, 0.18, ...], top_k=5)
for result in results:
    print(f"{result.id}: score={result.score:.3f}, {result.metadata}")

# Filter by metadata
results = store.search(query, top_k=10, filter=lambda m: m["category"] == "tech")

# Minimum score threshold
results = store.search(query, min_score=0.7)

# Persistence
store.save("vectors.json")
loaded = VectorStore.load("vectors.json")

# Batch operations
store.add_many([("id1", emb1, meta1), ("id2", emb2, meta2)])

API

Method	Description
add(id, embedding, metadata?)	Add a vector
add_many(items)	Batch add
search(query, top_k?, metric?, filter?, min_score?)	Similarity search
get(id)	Get entry by ID
delete(id)	Delete entry
update_metadata(id, metadata)	Update metadata
save(path)	Save to JSON
VectorStore.load(path)	Load from JSON
clear()	Remove all entries
ids()	List all IDs

Distance Metrics

• "cosine" (default) — cosine similarity
• "dot" — dot product

License

MIT