From Synthetic Data to Supercharged Models: Polite Guard

@IntelDevTools

Subscribe Now

Stay in the know on all things CODE. Updates are delivered to your inbox.

Sign Up

Overview

Polite Guard is the new open source project from Intel that uses synthetic data to develop a scalable natural language model for text classification. As the first politeness benchmark available to the developer community, Polite Guard empowers developers to build more robust and context-aware AI systems. 

Key Takeaways

This video walks through the creation process of Polite Guard, using Intel® Tiber™ AI Cloud and Intel® Gaudi® AI accelerators, and covers:

  • Streamlining dataset generation and fine-tuning for automation, efficiency, and quality
  • Hands-on exercises demonstrating techniques using:
    • PyTorch* Lightning for efficient training
    • ONNX* (Open Neural Network Exchange) for seamless model interoperability
    • Web Neural Network API (WebNN) and WebGPU for deploying models in web environments