From Zero to SOTA: Fine-Tuning BERT Using Hugging Face for World-Class Training Performance

Explore a comprehensive conference talk on fine-tuning BERT using Hugging Face for state-of-the-art training performance. Learn how Graphcore accelerates AI development through their Intelligence Processing Unit hardware and easy-to-integrate examples. Discover the implementation and optimization of BERT-Large for IPU systems, showcasing industry-leading performance results. Follow a step-by-step demonstration on accessing IPUs via Spell's Cloud MLOps platform, navigate through a BERT Fine-tuning notebook tutorial using the SQuADv1 dataset, and execute an inference question-answering task with the HuggingFace inference API. Gain insights into training transformer models faster using the Hugging Face Optimum toolkit, and understand the significance of BERT in industries undergoing AI transformation such as legal, banking and finance, and healthcare.

From Zero to SOTA: How to fine tune BERT using Hugging Face for world class training performance