Generating New Recipes using GPT-2

This 2 hour-long project will teach you how to preprocess a text data set that contains recipes and then split it into a training or validation set. You will learn how to use HuggingFace to fine-tune deep, generative models, and how to train such models on Google Colab. Finally, you will learn how GPT-2 can be used to create unique and realistic recipes using the above dataset. This project will teach you how to fine tune a large-scale model and the immense amount of resources required for it to learn. In this course, you will also learn about knowledge distillation. Note: This course is best for those who live in the North America region. We are currently working to offer the same experience in other areas.