Please use this identifier to cite or link to this item:
https://open.uns.ac.rs/handle/123456789/2073
Title: | IP core for efficient Zero-Run length compression of CNN feature maps | Authors: | Erdeljan, Andrea Vukobratović B. Struharik, Rastislav |
Issue Date: | 1-Jan-2018 | Journal: | Telfor Journal | Abstract: | © 2018 Academic Mind. Convolutional Neural Networks (CNNs) are becoming a fundamental tool for machine learning. High performance and energy efficiency are of great importance for deployments of CNNs in many embedded applications. Energy consumption during CNN processing is dominated by memory access and since large networks do not fit in on-chip storage, they require expensive DRAM access. This paper introduces an universal Output Stream Manager (OSM) which can be used to compress and format data coming from a CNN accelerator and reduce external memory access. The OSM exploits the sparsity of data and implements two Zero-Run Length encoding algorithms and can be easily reconfigured to optimize usage for different CNN layers. | URI: | https://open.uns.ac.rs/handle/123456789/2073 | ISSN: | 18213251 | DOI: | 10.5937/telfor1801044E |
Appears in Collections: | FTN Publikacije/Publications |
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.