In this course, we will create a custom Q&A bot powered by OpenAI! Along the way, you'll learn how OpenAI works and how to leverage its powerful language processing capabilities to build a functional Q&A bot that can provide insightful answers to your questions and impress your friends with its knowledge.
We'll start by creating an unsupervised machine-learning workflow to match the user's question to the relevant context in our dataset. We'll use that workflow to send a custom prompt that includes this context to an OpenAI text completion model. The output will be a custom response that is better aligned with our user's need for accurate data about recent events.
Overview
Syllabus
- Welcome
- In this lesson, you'll get an overview of our journey. You'll learn a bit about OpenAI's history and how we can use it. You'll also get a brief introduction to prompt engineering.
- Preparing a Dataset with Embeddings
- In this lesson, we'll prepare our dataset to make it accessible for machine learning. We'll find the data, load it, clean it up, and use embeddings to create numeric representations of our data.
- Finding Relevant Data
- In this lesson, we will discuss the differences between keyword and semantic searches. Then, we'll find the relevant data for our queries using cosine similarity.
- Querying a Completion Model with a Custom Text Prompt
- In this lesson, we'll get to the magic of AI! We'll write a custom prompt that includes the most relevant parts of our dataset and use that prompt to get a customized response.
- Project: Build Your Own Custom Chatbot
- For this project, you will use everything you learned in this course to create a custom chatbot using a dataset of your choice.
