Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
AI - Create Your Personal Document Assistant
Rating: 3.9 out of 5(4 ratings)
23 students

AI - Create Your Personal Document Assistant

Build Your Own Personal Document Assistant: Harnessing Llama 3.2, BGE Embeddings, and Qdrant Vector Database
Created byAdnan Waheed
Last updated 10/2024
English
English [Auto],

What you'll learn

  • Create personal chatbot with Llama3.2, Ollama, Langchain, Qdrant database
  • Load a LLM via docker
  • Load a vector database Qdrant via docker
  • Create a personal chatbot

Course content

2 sections • 18 lectures • 2h 9m total length

  • Introduction4:39

    Build a fully local personal document assistant that runs on your machine, using embeddings, a quadrant vector database in Docker, and an interactive chat to query your documents.

  • Code files0:05
  • The Game Plan3:45

    Build a streamlit app to upload and preview PDFs, convert them to embeddings in a quadrant vector database via docker, and run an llm chat bot locally with llama 3.2.

  • Create virtual environment and install dependencies5:00

    create and activate a python virtual environment, verify python 3.11.6, and install dependencies from requirements.txt, including streamlit, lang chain, llama, hugging face, and quadron for your document assistant.

  • Create a Streamlit App - Sidebar6:43

    Create a Streamlit app for a personal document assistant by building a sidebar with markdown, a home and chatbox dropdown, using session state in the app layout.

  • Create home page screen6:15

    Build the home page for the personal document assistant, featuring headings with icons, descriptive text, a bottom footer, and a dynamic title that switches with the chatbot view.

  • Upload a PDF Document8:57

    Build a Streamlit chat bot interface to upload a PDF document, create three expanders for upload document, embeddings, and chat with your document, and preview the file details.

  • Preview PDF document on the fly7:47

    Learn to preview a PDF uploaded to your chatbot by encoding the file in base64, embedding it in an iframe, and displaying the uploaded file name and size with Streamlit.

  • Initialize session states and store PDF file locally5:49

    Initialize session states in streamlit to track the pdf path, chat manager, and messages, then save the uploaded pdf locally as a temp file to enable embedding.

  • Load Qdrant vector database via Docker9:29

    Learn how to load the Qdrant vector database via Docker, install Docker, pull the quadrant image, run with port 6333, and store embeddings locally for your personal document assistant.

  • Create embeddings via BGE model12:42

    Create embeddings of an uploaded pdf using a Hugging Face open source model and store them in a quadrant vector database, triggered by a checkbox.

  • Process unstructured PDF, store embeddings in Qdrant10:37

    This lecture demonstrates processing unstructured pdfs to create embeddings with a BGP model and store them in a local Qdrant vector database, using a recursive text splitter.

  • Load llama3.2 model via Ollama4:22

    Load llama 3.2 via ollama to run a 3-billion-parameter model locally on your Mac or Windows PC, then build a retrieval-augmented chatbot that queries quadrant data.

  • Create a chatbot RAG7:20

    Load llama 3.23 billion parameters with quadrant vector database to enable chatbot development. Build a chatbot manager with LangChain, Hugging Face embeddings, prompt templates, and retrieval QA for RAG conversations.

  • Define PromptTemplate, Initiate Qdrant client/vector store10:10

    Define a prompt template to guide the language model with pdf context from the quadrant vector database, using embeddings and a retrieval augmented qa chain for accurate answers.

  • Create a chatbot3:59

    Initialize the chatbot manager after embedding completion and configure it with the session manager, llama 3.23.2, embeddings, 3 billion parameters, 0.7 temperature, localhost:63, vector db.

  • Setup chatbot with RAG20:54

    Set up a personal document assistant by embedding uploaded pdfs, initializing the chat box manager with llama 3.23 billion parameters, and querying the local quadrant database for answers.

Requirements

  • Basic Python programming knowledge
  • Desire to learn and excel more

Description

Managing multiple documents and finding the right information quickly can be a challenge. A personal document assistant simplifies this by allowing you to upload documents, ask questions, and get instant responses or summaries, making your work more efficient.

Benefits of Free Resources with Cutting-Edge Technology:

This course enables you to build a powerful system using free resources without compromising on advanced technology. By leveraging the BGE Embeddings model, Llama 3.2b, and the Qdrant vector database, you can run everything locally on your machine, ensuring both privacy and performance

In this course, we will build YOUR VERY OWN PERSONAL DOCUMENT ASSISTANT from scratch

Technologies Used:

  • Large Language Model: Llama 3.2b

  • Embeddings: BGE Embeddings

  • Vector Database: Qdrant (running locally within a Docker Container)

Features:

  • Personal: All technologies run locally on your own system.

  • Upload Documents: Easily upload your PDF documents.

  • Free Embeddings: Run embeddings on your documents with the FREE BGE Embedding model.

  • Chat: Interact with your documents via our intelligent chatbot. Ask questions, summarize documents, and much more.

Why Sign Up?

  • Personalized Learning: Hands-on exercises, real-world applications, and guided support.

  • Cutting-Edge Technology: Learn how to work with state-of-the-art models and databases, like Llama 3.2b and Qdrant.

  • Completely Local: Everything runs on your own system, no need to rely on external servers.

  • Interactive Chatbot: Build a functional chatbot that can interact with your documents and provide valuable insights in real-time.

  • Free Tools: Take advantage of the free BGE Embeddings model to enhance your assistant without any cost.

Add this to your portfolio AI PROJECTS!

Enroll NOW and create your very own personal document assistant system!

Who this course is for:

  • Anyone who want to explore the world of AI, LLM, ChatBot, Vector Database
  • Anyone who want to step into Vector Database world with practical learning
  • Data engineers, database administrators and data professionals curious about the emerging field of vector databases
  • Software developers interested in integrating vector databases into their applications.

Report abuse