Skip to main content

RAG Stores

Vector stores and storage backends for RAG systems including OpenSearch, Azure, and Vertex AI.

Overview​

The RAG stores system provides comprehensive vector storage capabilities with support for multiple backends and advanced features.

Core Features​

  • Multiple Backends: OpenSearch, Azure AI Search, Vertex AI, Pinecone
  • Hybrid Storage: Vector and metadata storage
  • Scalability: Horizontal scaling capabilities
  • Performance: Optimized for high-throughput operations
  • Integration: Seamless integration with RAG components

Usage Examples​

Basic Vector Store Setup​

from recoagent.rag.stores import OpenSearchStore

# Create OpenSearch store
vector_store = OpenSearchStore(
    host="localhost",
    port=9200,
    index_name="documents"
)

# Add documents
vector_store.add_documents([
    {"content": "Document content...", "metadata": {"title": "Doc 1"}},
    {"content": "Another document...", "metadata": {"title": "Doc 2"}}
])

# Search documents
results = vector_store.similarity_search("query", k=5)

Advanced Store Configuration​

from recoagent.rag.stores import AzureAISearchStore

# Create Azure AI Search store
azure_store = AzureAISearchStore(
    endpoint="https://your-search.search.windows.net",
    api_key="your-api-key",
    index_name="documents",
    vector_config={
        "dimensions": 1536,
        "similarity": "cosine"
    }
)

# Advanced search with filters
results = azure_store.search(
    query="machine learning",
    filters={"category": "AI", "date": "2024-01-01"},
    top_k=10
)

API Reference​

BaseVectorStore Methods​

add_documents(documents: List[Dict]) -> None​

Add documents to vector store

Parameters:

  • documents (List[Dict]): List of documents

similarity_search(query: str, k: int = 5) -> List[Document]​

Search for similar documents

Parameters:

  • query (str): Search query
  • k (int): Number of results

Returns: List of similar documents

See Also​