You’ve heard about ChatGPT’s conversational ability and how DALL-E can create images from a simple phrase. Now, you want to get your hands dirty training some state of the art (SOTA) deep learning models. We will use Jupyter notebooks to fine-tune an NLP model based on BERT to do sentiment analysis.

In this hands-on tutorial, we will learn about using HuggingFace models from pre-trained open-source checkpoints and adapting these models to our own specific tasks. We will see that using SOTA NLP and computer vision models has been made easier with a combination of HuggingFace and PyTorch.

At the end of this session, you will know how to fine-tune a large public pre-trained model to a particular task and have more confidence navigating the deep learning open source landscape.