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

Quantization is used to speed up execution time and save power when runnningDeep neural networks (DNNs) on edge devices like AI chips. To investigate theeffect of quantization, we need performing inference after quantizing theweights of DNN with 32-bit floating-point precision by a some bit width, andthen quantizing them back to 32-bit floating-point precision. This is becausethe DNN library can only handle floating-point numbers. However, the accuracyof the emulation does not provide accurate precision. We need accurateprecision to detect overflow in MAC operations or to verify the operation onedge de vices. We have developed PyParch, a DNN library that executes quantizedDNNs (QNNs) with exactly the same be havior as hardware. In this paper, wedescribe a new proposal and implementation of PyParch. As a result of theevaluation, the accuracy of QNNs with arbitrary bit widths can be estimated forla rge and complex DNNs such as YOLOv5, and the overflow can be detected. Weevaluated the overhead of the emulation time and found that it was 5.6 timesslower for QNN and 42  times slower for QNN with overflow detection compared to the normal DNNexecution time.

Development of Quantized DNN Library for Exact Hardware Emulation