Document Library

Reference architectures, white papers, and solutions briefs to help build and enhance your network infrastructure, at any level of deployment.

Intel® Deep Learning Boost (Intel® DL Boost) - Improve Inference Performance of Hugging Face BERT Base Model in Google Cloud Platform (GCP) Technology Guide

Last Updated: Apr 19, 2023

BERT is the best model to detect malicious and phishing websites/emails attacks as it provides very high accuracy. However, it takes longer inference time when compared to the traditional methods. This guide shows how to take advantage of Intel® AVX-512 Vector Neural Network Instructions (Intel® AVX-512 VNNI), oneDNN, and IPEX tool to boost AI Inference performance using Hugging Face BERT base model as an example. The evaluations were conducted on Google Cloud Platform* service (GCP) using three different hardware configurations.

Download PDF