Building End-to-End RAG Applications with Unstructured Data - Part 3: Docker Integration
The Machine Learning Engineer via YouTube
Overview
Learn to build the final component of an end-to-end RAG (Retrieval Augmented Generation) application in this third installment of a video series focused on interacting with unstructured data. Explore Docker containerization through detailed explanations of Dockerfile and docker-compose configurations. Master the integration of multiple technologies including Langchain for orchestration, various LLMs (Gemini Pro Fast, Microsoft Phi3.5 Mini, and LLama 3.2 3B hosted on Nvidia NIM), and different vector stores (Elastic, Choma, Faiss, and Vilmus). Implement embedding models from Gemini and Nvidia while utilizing Streamlit for the user interface and application server. Access the complete project through the provided GitHub repository to create a system capable of processing diverse unstructured data formats including audio, video, images, PDF, Excel, CSV, and HTML files.
Syllabus
RAG: E2E Application. Speak with your Unstructured Data Part 3 #machinelearning #datascience
Taught by
The Machine Learning Engineer