This project demonstrates how to use Feast to power a Retrieval-Augmented Generation (RAG) application.

In particular, this example expands on the basic RAG demo to show:

1. How to transform PDFs into text data with Docling that can be used by LLMs
2. How to use Milvus as a vector database to store and retrieve embeddings for RAG
3. How to transform PDFs with Docling during ingestion

• Online retrieval of features: Ensure real-time access to precomputed document embeddings and other structured data.
• Declarative feature definitions: Define feature views and entities in a Python file and empower Data Scientists to easily ship scalabe RAG applications with all of the existing benefits of Feast.
• Vector search: Leverage Feast’s integration with vector databases like Milvus to find relevant documents based on a similarity metric (e.g., cosine).
• Structured and unstructured context: Retrieve both embeddings and traditional features, injecting richer context into LLM prompts.
• Versioning and reusability: Collaborate across teams with discoverable, versioned feature transformations.

• data/: Contains the demo data, including Wikipedia summaries of cities with sentence embeddings stored in a Parquet file.
  • Note, you ahave to use the docling-demo.ipynb to construct the docling_samples.parquet file, the metadata_samples.parquet file are provided for you.
• example_repo.py: Defines the feature views and entity configurations for Feast.
• feature_store.yaml: Configures the offline and online stores (using local files and Milvus Lite in this demo).

The project has two main notebooks:

1. docling-demo.ipynb: Demonstrates how to use Docling to extract text from PDFs and store the text in a Parquet file.
2. docling-quickstart.ipynb: Shows how to use Feast to ingest the text data and store and retrieve it from the online store.