0% found this document useful (0 votes)
26 views

Mastering Generative AI (1)

The document outlines a two-day course on mastering Generative AI, covering topics such as Generative Adversarial Networks (GANs), Transformers, Large Language Models (LLMs), and Reinforcement Learning from Human Feedback (RLHF). Participants will engage in hands-on activities, including creating deep fake models, customizing LLMs, and building retrieval augmented generation (RAG) pipelines. The course is open to students from various backgrounds interested in advancing their careers in AI.

Uploaded by

myownmail687
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
26 views

Mastering Generative AI (1)

The document outlines a two-day course on mastering Generative AI, covering topics such as Generative Adversarial Networks (GANs), Transformers, Large Language Models (LLMs), and Reinforcement Learning from Human Feedback (RLHF). Participants will engage in hands-on activities, including creating deep fake models, customizing LLMs, and building retrieval augmented generation (RAG) pipelines. The course is open to students from various backgrounds interested in advancing their careers in AI.

Uploaded by

myownmail687
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

____________________________________________________________________________

Mastering Generative AI
(Exploring the Frontiers of Generative Models)

Course Content
Duration: 2 Days (14-16 HRS)
Eligibility: Can be joined by both Engineering and bio-medical and other students also who are
interested in making their career and want to update in AI. (Open to All)

Introduction to Generative AI:


Get to know about what is GenAI –
​ Difference between AI, Machine Learning & Deep Learning
​ Difference between Discriminative AI vs Generative AI
​ Tools involved in GenAI
​ Know about privileged GenAI tools and Models.
​ Effective Road Map to explore the GenAI Model.

Generative Architecture

​ Introduction to Generative Adversarial Network (GAN) Architecture


​ Know about the Generator & Discriminator Models
​ How GAN architecture uses for Deep Fake Model
​ Difference between Variational Auto Encoders (VAEs) vs GAN architecture
​ Hands-on: Development of Generative Images using GAN Complex Architecture
​ Hands-on: Create your own “Deep Fake Model” using low-level GAN Structure.
​ Deep insights of GAN applications & drawbacks of it.

_________________________________________________________________________________________________________
Techobytes Technologies
Contact: 91+6397685589 | [email protected]
Address: BS-915, Techobytes Technologies, Diamond Galaxy Plaza, Noida Ext-201306
____________________________________________________________________________

Attention Is All You Need – Transformers & LLMs

​ What is Transformer? How the architecture looks like?


​ Text Generation before transformers

​ Know about Attention Model vs Multi-Head Attention Model vs Scaled Dot- Product Attention
​ Difference between Encoder & Decoder Architecture.
​ Difference between Batch vs Layer Normalization
​ Step-into the most famous transformer community – Hugging Face
​ Text Generation after transformers
​ Introduction to Large Language Models (LLMs)
​ Popular LLMs: BERT, BART, DistilBert
​ Hands-on: Conversation AI using pre-trained LLM Model
​ Hands-on: Image Caption Generator using pre-trained LLM Model
​ Hands-on: Customized & Fine-tuning Question & Answering BERT Model
​ Hands-on: Customized & Fine-tuning DistilBert Model

Mastering Generative AI (Exploring the Frontiers of Generative Models)


​ Learn about GPT (Generative Pre-trained Model )
​ Fine-Tuning the LLM Model on a specific task & Model Evaluation
​ Parameter Efficient Fine-Tuning (PEFT)
​Vision Transformer – ViT Model with API interface

​ Learn about Vision Transformer Architecture and its variable components.
​ Understand about Image Vectorization, Patches extraction & Positional Encoding.
​ Hands-on: Pre-trained Google ViT Model for Image Classification
​ Hands-on: Customized ViT architecture (RAW Structure for ViT Base)
​ Hands-on: Create your own ViT Multi-class classification and deploy the model into
​ HuggingFace Hug using API Interface.
​ Git workflows
​ Git cheat sheet
_________________________________________________________________________________________________________
Techobytes Technologies
Contact: 91+6397685589 | [email protected]
Address: BS-915, Techobytes Technologies, Diamond Galaxy Plaza, Noida Ext-201306
____________________________________________________________________________


​Reinforcement Learning From Human Feedback:

​ Deep Insight of RLHF


​ RLHF: Obtaining feedback from humans
​ RLHF: Reward model
​ RLHF: Fine-tuning with reinforcement learning

​ Relative Sync of ChatGPT with RLHF

​GenAI – LangChain Framework:

​ What is LangChain?
​ Usage of LangChain Framework? How it works?
​ Integrating the LangChain Framework with LLM Model.
​ Understand about the Langchain pipeline end points, Chains & Prompt Template.
​ Hands-on: Create the simple LLM Langchain using the HuggingFace API Interface
​ Hands-on: Create the Langchain framework including with promptTemplate,
​AutoTokenizer & LLM Chain interfacing with HuggingFace API Interface

GenAI – Retrieval Augmented Generation (RAG)



​ Explore the different tools in LangChain and initialise an agent that uses the tools to
​ read different types of files or data present in the company database
​ Build the backend for the system using Vectorstore options present in LangChain
​ Divide the documents into chunks and apply the LLM to create the embeddings.
​ Extract entity for the chunks of document and store them in the Vectorstore

​ Construct the Search Index and Entity Store and create a functionality to update it with
​ every question that the user asks

​ Use the Chain functionality of LangChain to connect all the components



_________________________________________________________________________________________________________
Techobytes Technologies
Contact: 91+6397685589 | [email protected]
Address: BS-915, Techobytes Technologies, Diamond Galaxy Plaza, Noida Ext-201306
____________________________________________________________________________

​ Hands-on: Create your simple RAG pipeline with FAISS Vector Store similarity
​ search using OpenAI API or HuggingFace End-point API Interface.

​ Hands-on: Create your own Retrieval QA End-to-End RAG pipeline using
​ OpenAI/HuggingFace API.

​GenAI Advanced- LangGraph & VectorDB/ChromaDB

​ uild an agent from scratch, and understand the division of tasks between the LLM and
B
​the code around the LLM.
​ Integrate agentic search capabilities to enhance agent knowledge and performance.
​ Learn how agentic search retrieves multiple answers in a predictable format, unlike
​traditional search engines that return links.
​ Incorporate human-in-the-loop into agent systems

​ Hands-on: Create your own Retrieval pipeline using ChromaDB to store the pdf
​embedding chunks.

​Add-on: Hugging Spaces- Knowledge to deploy
​ How to use the Llama & other popular LLM models in HuggingFace Spaces.
​ How to create the new workspace and initializing your repository?

​ Hands-on: Build and deploy your own GenAI Application in Hugging Face Spaces
​using Streamlit.

_________________________________________________________________________________________________________
Techobytes Technologies
Contact: 91+6397685589 | [email protected]
Address: BS-915, Techobytes Technologies, Diamond Galaxy Plaza, Noida Ext-201306

You might also like