Openai, Gpt, Chatgpt And Dall-E Masterclass

Everything from No-Code, through to fine-tuning & embedding with code examples in multiple programming languages.

What you'll learn

No-Code examples using OpenAI playground

Techniques to create, expand, rewrite, and summarize text for creative writing, articles and blogs using GPT

In depth Prompt Engineering with examples and Tips/Tricks

Understand what every fine tuning parameter does with recommended values

How to write code to call GPT and OpenAI using several different programming languages

How to generate, document, and explain code and SQL in plain english using CODEX

Deep dive into creating and uploading Fine Tuning sets to train GPT with your own data

How to use Embedding to search large documents and ask questions related to its content

How to use Embedding for Clustering and Classification to find hidden patterns

Create and modify images using DALL-E

Write your own chatbot using GPT

Best Safety Practices and Cost Saving Tips

Requirements

Some programming knowledge may help - but is not required

Most examples can be done in the OpenAI playground

Description

RequirementsBeginner and Intermediate level: NoneAdvanced level content: Programming skills in any languageDescriptionWelcome to the Masterclass for GPT, DALL-E and ChatGPT.Ever since OpenAI arrived on the scene, access to a trained AI has become accessible to everyone.GPT allows you to ask a chatbot to complete tasks, and to answer questionsFine-tuning allows you to change the way the AI respondsEmbedding allows you to use your own knowledge baseDall-E allows you to generate images from textIn this course, you will learn very practical skills for using GPT. The skills can be used in the OpenAI playground or in programming code.The course is split into 6 major parts:Introduction and Prompt EngineeringWriting code and calling the APIDall-E and CodexFine-tuning EmbeddingWriting and managing ChatbotsPART 1: Introduction and Prompt EngineeringIn this section, you will learn how to get started with GPT. We also explain, design and use different types of prompts. At the end of the prompt engineering section, we give you two recipes you can use to get consistent results.We'll see how GPT works, including:what tokens arehigh level overviews of GPT, Codex and Dall-Elive examples of using the playgroundways to improve prompts to get better resultshow to use templating to create reliable promptshow to use context to introduce new knowledgeThis is already very practical.We cover everything from "zero-shot" queries, through to multi-shot and advanced template prompts.An entire section of the course is dedicated to creative writing for blogs, books and articles.PART 2: Writing Code to use the APIIn this section, you will learn how to write code to call GPT and the OpenAI API. Most of the examples are written in Python, but they are equally usable in C#, Typescript, Javascript, Node.js, PHP, Powershell and many other languages.There are examples and explanations of how to adapt the code for each of these languages. We introduce various coding libraries and point out issues that may be specific to each language.Once you start writing code, we go into all of the possible parameters and how they can change the way GPT works. We explain what they do, and ideal values for different situations.As well as covering the completions endpoint, we also talk about the edit or instruction end point.PART 3: CodeX and Dall-EIn this section, you will learn how you can use CodeX to generate, debug, and document code. We also give an example of how you can use CodeX to create unit tests for your functions.You will find out how you can tell CodeX about new API calls and functions that are not part of its standard learning. We also use the edit endpoint to create an entire function from scratch.Because CodeX generates code, we discuss the safety of using the code it generates and highlight several serious issues that you need to consider if you intend to use the code in production. We point out possible vulnerabilities and ways hackers can exploit the code it generates. When it comes to Dall-E we walk through practical code examples to creating or edit images. We provide examples in multiple programming languages and explain how to handle images in memory and on your file systemPART 4: Fine TuningIn this section we explain how you can fine-tune GPT. We explain the benefits and the difficulties. There are plenty of training examples and strategies you can use. We walk you through creating a set from scratch and uploading it OpenAI using your own code.This followed by  an entire section explaining how you can use GPT to create its own trainings sets and how to get GPT to check and improve its own outputs.To round out this section of the course, we go though all of the fine-tuning parameters. We explain how you can change settings to adjust the impact the rules have on the base training.Part 5: Training on Large Text Documents using EmbeddingEveryone wants to know how you can train GPT on large text documents. This section of the course explains how to take a text document and use it to answer questions using GPT. Embedding vectors are explained in great detail. We also explain the theory behind them so you know what GPT is doing. We talk about breaking up large text documents, creating embeddings, and searching the results.We also talk about creating highly accurate classifiers and using clustering to find hidden patterns within documents and text.Part 6: ChatbotsWho doesn't want their own chatbot? We explain how you can use GPT as a fully functional chatbot. We explain about personas and give you a recipe you can use to give your chatbot a personality and keep it on track.To keep your chatbot on task and to give it persistent memory, we explain how you can use embedding to enhance the functionality and provide unique knowledge that is not part of the base GPT training.UNIQUE FEATURESEvery line of code explained in detail - email me any time if you disagreeNo wasted time "typing" on the keyboard like other courses. Instead we show finished code examples and prompts with detailed explanations that you can apply to your own use cases.Thank you for reading and I hope to see you soon!Who this course is for:Anyone who wants to master GPT and OpenAIAnyone who loves deep Natural Language ProcessingAnyone who wants to create their own chatbots or products using OpenAI and GPT

Overview

Section 1: What is OpenAI?

Lecture 1 Introduction

Lecture 2 Quick Start Code Example

Lecture 3 Text Completion High Level Overview

Lecture 4 CODEX High Level Overview

Lecture 5 What are Tokens?

Lecture 6 DALL-E High Level Overview

Lecture 7 Naming Conventions I Use in the Course

Section 2: No-Code Testing Using the OpenAI Playground

Lecture 8 Examples of Using the OpenAI Playground for No-Code Testing of Prompts

Section 3: Prompt Engineering

Lecture 9 Introduction to GPT Prompt Engineering

Lecture 10 Improving GPT Prompts to get Better Responses

Lecture 11 Recipe 1 : GPT Templating

Lecture 12 Recipe 1 : GPT Template Examples

Lecture 13 Recipe 2 : Including Context or Knowledge in GPT Prompts

Section 4: Writing Code

Lecture 14 Getting Started

Lecture 15 The Anatomy of a Request and Response

Lecture 16 Testing in the Playground

Lecture 17 Code Examples in a Range of Programming Languages

Lecture 18 PHP Example without a Library

Lecture 19 Powershell Example

Lecture 20 Using Javascript Fetch and Cross Site Origin Issues

Section 5: The Moderation Endpoint

Lecture 21 Calling the Moderation Endpoint to Stay Safe and Interpreting the Results

Section 6: Tweaking Prompt Parameters

Lecture 22 Introduction to Completion Parameters

Lecture 23 Quick Revision of a Code Example to call the API

Lecture 24 max_tokens

Lecture 25 top_p

Lecture 26 temperature

Lecture 27 n

Lecture 28 stop

Lecture 29 best_of

Lecture 30 suffix

Lecture 31 echo

Lecture 32 user

Section 7: Adjusting Probabilities and Avoiding Repetition

Lecture 33 Introduction to Token Probabilities

Lecture 34 View Token Probabilities with the logprobs Parameter

Lecture 35 Using the Tokenizer to Get Token ID's and Adjust Token Probabilities

Lecture 36 Adjusting a Token using logit_bias

Lecture 37 Avoiding Repetition with Penalty Parameters

Section 8: Using the Edits Endpoint with Input and Instruction

Lecture 38 An Example of Using Edit Instead of Completion

Section 9: DALL-E

Lecture 39 Introduction to DALL-E

Lecture 40 Creating Images using DALL-E from a Text Prompt

Lecture 41 Modifying Existing Images using DALL-E and Mask Files

Lecture 42 Generating Variations of Images using DALL-E

Lecture 43 Tips for handling Images with Different Programming Languages

Lecture 44 Image Moderation and Avoiding User Abuse

Section 10: CODEX - Learn By Example

Lecture 45 Introduction to Codex and the Code Models

Lecture 46 Example : Simple Comments to Python Code

Lecture 47 Examples : Comments to SQL

Lecture 48 Example : Explain What a Function or Code Does

Lecture 49 Checkpoint : Best Practices

Lecture 50 Example : Comprehensive Prompt for a Complex Task

Lecture 51 Example : Generate Unit Tests

Lecture 52 Example : Finding and Fixing Bugs in Code

Lecture 53 Example : Convert Between Programming Languages

Lecture 54 Example : Work Out Time Complexity for a Function

Lecture 55 Example : Teaching New API Definitions

Lecture 56 Example : Step by Step Construction of a Function Using Inputs and Instructions

Lecture 57 IMPORTANT : Writing Safe Code Using CodeX (Tips and Warnings)

Section 11: Fine-Tuning GPT

Lecture 58 What is Fine Tuning?

Lecture 59 The Anatomy of a JSONL file

Lecture 60 Training to Determine Sentiment : Analyze Tweets

Lecture 61 Training to Create a True/False Checker : Matching Company Names and Slogans

Lecture 62 Training for Classification : Example Based on Email Contents

Lecture 63 Training to Summarize : Write Engaging Sales Copy from a Wikipedia Description

Lecture 64 Training to Expand : Write Sales Copy from Properties of an Item

Lecture 65 Training to Extract : Pull Entities from Emails or Text

Lecture 66 Fine Tuning a Chat Bot

Lecture 67 Fine Tuning Based on Text from Books and Documents

Lecture 68 Analyzing the Effectiveness of the Training

Lecture 69 Tips and Tricks

Section 12: Generating Fine Tuning Data Sets

Lecture 70 Scraping Data

Lecture 71 Get GPT to Generate it's own Training Data

Lecture 72 Get GPT to Check and Improve it's own Output

Section 13: Uploading Fine-Tuning Files and Tweaking Parameters

Lecture 73 Upload and Process the Fine-Tuning File

Lecture 74 Errors While Uploading

Lecture 75 batch_size and n_epochs

Lecture 76 learning_rate_multiplier

Lecture 77 Find and Use the Fine-Tuned Model

Section 14: Embedding, Searching and Integrating Large Documents and Text

Lecture 78 Part 1 : What are Embeddings

Lecture 79 Part 1 : The ADA 002 Embedding Engine

Lecture 80 Part 1 : Generate Your First Embedding Vector

Lecture 81 Part 1 : The Code Libraries You Will Need to Use

Lecture 82 Part 2 : Checking Text for Similarity

Lecture 83 Part 3 : Introduction to Embedding

Lecture 84 Part 3 : Combining Data to Create the Source Text

Lecture 85 Part 4 : Cleaning the Data So We Don't Break GPT

Lecture 86 Part 4 : Creating a Function to Calculate the Embedding Vectors

Lecture 87 Part 4 : Using the Function to Add Embedding to our Data Source

Lecture 88 Part 5 : Putting it to Work - Loading the Data Source

Lecture 89 Part 5 : The Final Step - Doing a Semantic Search

Lecture 90 Integrating Alternative Embedding Libraries with GPT - Word2vec and Pinecone

Section 15: Classifying Data using Embedding

Lecture 91 Classifying Text Using Embedding

Lecture 92 Testing the Accuracy of the Classifier

Section 16: Clustering Data Using Embedding

Lecture 93 How to Find Hidden Patterns in Data using Embedding and Clusters

Lecture 94 Final Summary of Embedding

Section 17: Using GPT for Creative Writing

Lecture 95 Creating an Article in Stages - Brainstorming and Generating Paragraphs

Lecture 96 Expanding Existing Text

Lecture 97 Summarizing Text

Lecture 98 Extracting Facts

Lecture 99 Rewriting Existing Articles, Blog Posts or Information

Section 18: Creating a Chatbot with GPT

Lecture 100 The Back and Forth Flow of Chatbots

Lecture 101 Tips for Training a Chatbot for Knowledge

Lecture 102 Using Embedding to Train a Chatbot

Lecture 103 Ring Fencing the Chatbot to Avoid Abuse

Lecture 104 Adding Persistence Memory to a Chatbot

Lecture 105 A Recipe to Give Your Chatbot a Persona

Section 19: Usage Policy and Best Safety Practices

Lecture 106 Staying Safe - Usage Policies from the Developers Point of View

Lecture 107 Staying Safe - Best Coding Practices

Lecture 108 Staying Safe - API Keys and Avoiding Prompt Injection

Lecture 109 Cost Saving Tips

Anyone interested in utilizing OpenAI and GPT for artificial intelligence