The Complete Openai And Gpt Course - Build A Q&A Chatbot

Master OpenAI's GPT, Semantic Search, Embeddings, and Q&A to build a Financial Assistant. Beginner friendly.

What you'll learn

The foundations of GPT and generative text - Large Language Models (LLM), Prompt Engineering

Receiver Augmented Generation (RAG) for Question Answering - its use cases and challenges, and real world implementation

Finetuning GPT models and their best practices, when and when not to fine tune.

Best practice strategies for troubleshooting issues with OpenAI APIs

Semantic Search - theory and Implementation

Vector databases, Pinecone - how they work, code samples

How to choose the right GPT model for completion and classification tasks

Understand how to use OpenAI’s APIs and their production best practices

Tackling the LLM hallucination problem - what the problem is, and specific strategies to mitigate it.

Requirements

Prior exposure to Python and Pandas. You don’t need to be an experienced Python and Pandas developer, but the ability to follow along and understand syntax is needed.

Github and Google accounts (free)

Description

Note: This course assumes that you have gotten the basics of Python and Pandas down. You don’t need to be an experienced Python and Pandas developer, but the ability to follow along and understand syntax is needed.Take your AI development skills to the next level with this course!In this course, you will learn how to build an AI assistant powered by OpenAI's GPT technology, HuggingFace, and Streamlit. In addition, you will learn the foundational concepts of GPT and generative AI, such as Large Language Models, Prompt Engineering, Semantic Search, Finetuning, and more. You will also understand how to use OpenAI’s APIs and their best practices, with real world code samples.Unlike other courses, you will learn by doing. You will start with a blank app, and add features one at a time. Before adding a new feature, you will learn just enough theory to confidently build your app.You will get all the code samples, including Google colab notebooks, and access to the Q&A forum if you get stuck. You don’t need a powerful PC or Mac that has GPUs to take this course. By the end of the course, you will be able to deploy and create your first app using OpenAI’s technology, and be confident about the theoretical knowledge behind this technology. So sign up today and start building your AI powered app!What you will learn:Creating an AI chatbot with StreamlitIntentClassifiers - what they are, how to build it.Prompt Engineering: different ways of crafting the perfect promptHow to evaluate and choose the best promptThe concept of word embeddingsHow to use word embeddings to quantify semantic similarityHow to use a vector database to store word embeddingsHow to create a search engine that searches based on word embeddingsHow to perform entity resolution for documentsSentiment extraction using GPTHow to clean a finance dataset for use in a semantic searchHow to embed finance documents and upload them to a vector databaseHow to use a language model to generate answers to questionsHow to use fine-tuning to ensure the language model does not hallucinateHow to deploy a Q&A bot and a custom action system.

Overview

Section 1: Introduction

Lecture 1 Introduction

Lecture 2 Course Logistics and Important Announcements

Lecture 3 What We Are Building & Problem Statement

Lecture 4 FAQ

Lecture 5 Important Disclaimers

Section 2: Project 0: Create a ChatGPT Clone with Python and Streamlit

Lecture 6 Course Setup

Lecture 7 Course Project Solutions

Lecture 8 Building a ChatGPT Clone in 50 lines of Code

Lecture 9 Integrating OpenAI

Section 3: GPT3, Prompt Engineering, and LLMs

Lecture 10 ChatGPT, GPT3, InstructGPT - How They Work

Lecture 11 Prompt Engineering and Advanced GPT Parameters

Lecture 12 Why GPT Disrupted AI Industry

Section 4: Project 1: Intent Classifier

Lecture 13 IntentClassifier - What It is, Why It's Important

Lecture 14 Prompts for Classification Problems (Notebook)

Lecture 15 Evaluation GPT3.5 for Classification (Notebook)

Lecture 16 Integrate Intent Classifier into the App

Section 5: Limits of GPT - What It Can't Do

Lecture 17 Limitations of GPT - Knowledge Cutoff, Data Gaps, Token Limits

Lecture 18 Limits of GPT - Reasoning, Chain of Thought Prompting

Section 6: Project 2: Semantic Search and Retrievers

Lecture 19 Semantic Search Based Retrieval

Lecture 20 Word and Sentence Embeddings (Notebook)

Lecture 21 Semantic Search (Notebook)

Lecture 22 Vector Databases, Pinecone, Nearest Neighbor Search

Lecture 23 Integrating News Article Retriever into App

Section 7: Project 3: Retriever Augmented Question Answering and Fine Tuninng

Lecture 24 Question Answering with GPT, and Finetuning GPT Models

Lecture 25 Question Answering, Strategies for Handling Hallucinations (Notebook)

Lecture 26 Question Answering and Finetuning GPT (Notebook)

Lecture 27 Generative Labeling, Finetuning GPT, Model Evaluation (Notebook)

Lecture 28 App Integration

Section 8: Project 4: Summarization, External System Integration

Lecture 29 Document Summarization with GPT

Lecture 30 Summarization with GPT (Notebook)

Lecture 31 Adding Real Time Financial Charts (Notebook)

Lecture 32 App Integration

Lecture 33 Deployment

Python developers with some Pandas experience who are eager to build their first AI app using GPT library