In this 2 hour long project, you will learn how to preprocess a text dataset comprising recipes, and split it into a training and validation set. You will learn how to use the HuggingFace library to fine-tune a deep, generative model, and specifically how to train such a model on Google Colab. Finally, you will learn how to use GPT-2 effectively to create realistic and unique recipes from lists of ingredients based on the aforementioned dataset. This project aims to teach you how to fine-tune a large-scale model, and the sheer magnitude of resources it takes for these models to learn. You will also learn about knowledge distillation and its efficacy in use cases such as this one.
Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.