Installation

The package can be installed with pip or uv:

pip install omop-graph
# or
uv pip install omop-graph

Embedding and RAG support (optional, recommended)

Tip

Installing with [emb] is recommended. Without it the KnowledgeGraph operates in text-only mode and all embedding-based similarity scoring is disabled.

pip install "omop-graph[emb]"
# or
uv pip install "omop-graph[emb]"

This pulls in omop-emb with its default sqlite-vec backend, a file-based vector store that requires no external database server. It still needs a configured vector store, model, and provider (via oa-configurator) before use; see KnowledgeGraph: Embedding Configuration.

It enables:

• vector similarity search over OMOP concepts
• embedding-weighted grounding scores
• on-the-fly embedding computation for un-indexed concepts

Scaling up: pgvector or FAISS

For larger deployments or approximate-nearest-neighbour acceleration, install the corresponding extra instead of (or alongside) [emb]:

Extra	What it adds
omop-graph[emb]	sqlite-vec backend (default, no external server needed)
omop-graph[pgvector]	PostgreSQL/pgvector backend
omop-graph[faiss-cpu]	FAISS sidecar for fast approximate search

These can be combined:

pip install "omop-graph[pgvector,faiss-cpu]"

omop-emb documentation

Backend configuration, CLI reference, and index management are covered in the omop-emb documentation. The [emb] extra mirrors the base omop-emb install; [pgvector] and [faiss-cpu] mirror omop-emb[pgvector] and omop-emb[faiss-cpu] respectively.