English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In the past few years, more and more AI applications have been applied toedge devices. However, models trained by data scientists with machine learningframeworks, such as PyTorch or TensorFlow, can not be seamlessly executed onedge. In this paper, we develop an end-to-end code generator parsing apre-trained model to C source libraries for the backend using MicroTVM, amachine learning compiler framework extension addressing inference on baremetal devices. An analysis shows that specific compute-intensive operators canbe easily offloaded to the dedicated accelerator with a Universal ModularAccelerator (UMA) interface, while others are processed in the CPU cores. Byusing the automatically generated ahead-of-time C runtime, we conduct a handgesture recognition experiment on an ARM Cortex M4F core.

Deploying Machine Learning Models to Ahead-of-Time Runtime on Edge Using MicroTVM