Efficient inference on IMG Series4 NNAs
February 6, 2023
Research into neural network architectures generally prioritises accuracy over efficiency. Certain papers have investigated efficiency (Tan and Le 2020) (Sandler, et al. 2018), but quite often this is with CPU- or GPU-based rather than accelerator-based inference in mind.
In this original work from Imagination’s AI Research team, many well-known classification networks trained on ImageNet are evaluated. We are not interested in accuracy or cost in their own right, but rather in efficiency, which is a combination of the two. In other words, we want networks that get high accuracy on our IMG Series4 NNAs at as low a cost as possible. We cover:
- identifying ImageNet classification network architectures that give the best accuracy/performance trade-offs on our Series4 NNAs.
- reducing cost dramatically using quantisation-aware training (QAT) and low-precision weights without affecting accuracy.
To read the full article, click here
Related Semiconductor IP
Related Blogs
- Deep learning inference performance on the Yitian 710
- ChiPy®: Bridge Neural Networks and C++ on Silicon — Full Inference Pipelines with Zero CPU Round-Trips
- Introducing Cortex-A35: ARM's Most Efficient Application Processor
- Energy Efficient Designs with MIPI
Latest Blogs
- Designing for What Comes Next in AI
- Secure boot on Efinix Titanium Ti60: Getting started with KiviBoot-Tiny
- How Long Does It Take to Port a New AI Model? Minutes Now, and Falling
- PCI Express Security and TEE Integration, Part 1 - Establishing Trust
- World Models at the Edge, in Real Time: V-JEPA 2.1 on the Quadric GPNPU