Generative AI Specialist (Batch-02)

Wishlist Share
Share Course
Page Link
Share On Social Media

What Will You Learn?

  • Master the creation of advanced generative AI applications using the Langchain framework and Huggingface's cutting-edge models.
  • Understand the architecture and design patterns for building robust and scalable generative AI systems.
  • Gain practical experience in deploying generative AI models across various environments, including cloud platforms and on-premise servers.
  • Explore deployment strategies that ensure scalability, reliability, and optimal performance of AI applications.
  • Develop Retrieval-Augmented Generation (RAG) pipelines to boost the accuracy and efficiency of generative models by integrating retrieval mechanisms.
  • Seamlessly incorporate Huggingface's pre-trained models into Langchain applications to leverage their powerful NLP capabilities.
  • Customize and fine-tune Huggingface models to meet specific application needs and use cases.
  • Engage in real-world projects demonstrating Generative AI applications in domains such as chatbots, content generation, and data augmentation.

Course Content

Introductory Class – Generative AI: Overview & Future Prospects | Career Opportunities in AI | Training Modules Breakdown | Instructor Introduction | Q&A Session

  • Demo Session
    01:58:25

Class 01 – Python Basics: Punctuations, Operators, Brackets & Logic | Syntax & Semantics | Conditional Statements | Variables & Types | Naming Conventions | Common Errors & Examples

Class 02 – Data Types (Basic & Advanced: List, Tuple, Set, Dict) | Operators (Arithmetic, Comparison, Logical) | Conditional Statements (IF, ELIF, ELSE) | Practical: Employee Bonus Calculator

Class 03 – Python Basics: Syntax, Variables, Data Types | Conditional Statements (If, Else, Elif, Nested) | Logical Operators

Class 04 – Conditional Statements Recap | Loops in Python (For, While, Nested, Break, Continue, Pass) | Introduction to Lists

Class 05 – Functions in Python (Definition, Calling, Arguments) | Lambda Functions | Map & Filter Functions

Class 06 – Python Dictionaries (Key-Value Pairs, Access/Modify Data, Methods) | Functions (Definition, Calling, Parameters, Return, Scope)

Class 07 – Object-Oriented Programming (OOP): Classes, Objects, Instance Methods | Single Inheritance

Class 08 – Object-Oriented Programming (OOP): Multiple Inheritance | Polymorphism (Method Overriding, Abstract Methods) | Encapsulation (Getters & Setters) | Abstraction | Magic Methods

Class 09 – Streamlit & ML Deployment: VS Code Setup | Streamlit Basics (Title, Text, DataFrames, Widgets) | Upload/Download Features | ML Classification App (Random Forest on Iris Dataset)

Class 10 – Intro to NLP | ANN & RNN Concepts | RNN Forward & Backward Propagation | RNN Limitations | Project Discussion

Class 11 – ANN Project: Dataset Overview | Feature Engineering (Label Encoding, One-Hot Encoding, Pickle Files, Train-Test Split, Scaling) | Model Training

Class 12 – ANN Project: Model Training (Dense, Optimizer, Learning Rate, Early Stopping) | TensorBoard Visualization (Val Loss & Accuracy) | Predicting Custom Inputs | Deploying Streamlit WebApp Locally

ChatGPT said: Class 13 – Recurrent Neural Networks (RNN): LSTM Architecture | Forget Gate | Input Gate | Candidate Memory | Output Gate

Class 14 – LSTM RNN Training Process | GRU Architecture Explained | End-to-End Project: Data Collection to Deployment

Class 15 – Bidirectional RNN Architecture | Seq2Seq Encoder-Decoder Model | Limitations of Seq2Seq | Introduction to Attention Mechanism

Class 16 – Attention Mechanism | Transformers Overview | Intro to Generative AI & LLMs | Understanding Key Research Papers

Class 17 – Generative AI Overview | AI vs ML vs DL | LLMs (ChatGPT, LLaMA 3) Training Process | Evolution of AI Models | LangChain Ecosystem Deep Dive

Class 18 – LangChain Implementation | PromptTemplates, OutputParsers, and LCEL Chaining | Human/System Messages | Chatbot with Conversational History | Model Setup, Invocation & Environment Connection

Class 19 – Project (Part 01): Conversational Q&A Chatbot App | Session IDs & Message History | Managing Multi-User Conversations | Context Retention & Query Storage

Class 20 – Project (Part 02): Conversational Q&A Chatbot App | Session Management & Context Retention | Multi-User Handling | Query Storage & Retrieval

Class 21 – Project (Part 01): RAG-Based Document Q&A | GROQ API Integration | Hugging Face Embeddings | Document Upload & Preprocessing | Retriever-Reader Pipeline | Contextual Querying

Class 22 – Project (Part 02): RAG-Based Document Q&A | GROQ API Integration | Hugging Face Embeddings | Document Upload & Preprocessing | Retriever-Reader Pipeline | Contextual Querying

Class 23 – Tools & Agents in LangChain | Defining Tools as External Resources | Introducing Agents for Tool Usage Decisions | Enhancing Search with Real-Time Data | Future Implementations Overview

Class 24 – Tool Creation & Custom Tools | Preparing for Agents | Combining LLM with Tool List | Creating Dynamic Agents | AgentExecutor Integration | Streamlit UI for User Queries & Groq API Key | Defining Multiple Tools with Dynamic Selection | Real-Time Chain of Thought Reasoning | Error Handling & Loop Prevention

Class 25 – Generative AI Project: Chat with SQL DB via LangChain SQL Toolkit & Agents | Auto SQL Query Generation | SQLite DB & Student Table Setup | Connected DB to LangChain Agent | MySQL Workbench Installation | Sample Data Creation & Verification | Streamlit Integration for Natural Language SQL Queries

Class 26 – Generative AI Project: Chat with SQL DB via LangChain & Streamlit | Built Web App for SQLite/MySQL | Collected Groq API Key & DB Credentials | Created Agents for Dataset Querying | Integrated DB Choice with LangChain Agent | Displayed Queries & Results in Streamlit Chat

Class 27 – Chat with SQL DB Application (Developed End-to-End Streamlit App for SQLite & MySQL | Implemented DB Configuration | Integrated User DB Choice with LangChain Agents | Displayed Queries & Results in Streamlit | Tested Auto-Generated SQL) | URL Summarization Project (Overview & UI Design | YouTube & Webpage Summarization | Text Chunking & Data Handling)

Class 28 – End-to-End Deployment of YouTube and Website Summarization Bot | Setup and Imports: Installed and Explained Streamlit, Validators, LangChain, YouTube-Transcript-API | UI Layout: Configured Page Title and Icon, Sidebar with Usage Steps, Secure API Key Input, Summary Length Slider | Input Validation: Collected API Key and URL, In-Body Warning for Missing Key, URL Format Validation | Content Loading: Used YoutubeLoader with Error Handling for Disabled Captions, UnstructuredURLLoader for Webpages | LLM Integration: Initialized ChatGroq with API Key, Built Dynamic PromptTemplate, Ran Load_Summarize_Chain | UX Feedback: Added Spinner During Fetch and Summarize, Displayed Success and Error Messages | Clean Code Practices: Added Inline Comments, Modularized Code into Validation, Loading, LLM, and UI Sections for Readability

Class 29 – Project 11: Fine-Tuning & Quantization | Quantization Concepts & Precision Formats | Post-Training & Quantization-Aware Training | LoRA & Quantized LoRA (QLoRA) | Calibration & Low-Rank Approximations

Class 30 –

Class 31 – Project 11: Fine-Tuning Techniques for Efficient Deployment | Quantization (FP-32, FP-16, INT-8) | Calibration & Quantization-Aware Training (QAT) | Low-Rank Adaptation (LoRA) | Quantized LoRA (QLoRA) for Compact Fine-Tuning

Class 32 –