- Step1: Clone the Advanced_RAG repository from GitHub.
- Step2: Install required dependencies using pip install -r requirements.txt.
- Step3: Configure your vector store (e.g., FAISS, Pinecone) in the config file.
- Step4: Load and index your documents using the provided ingestion scripts.
- Step5: Customize the retriever and LLM settings in the pipeline.
- Step6: Run the RAG pipeline script to query and generate responses.
- Step7: Evaluate and tune parameters using built-in evaluation modules.