Por favor, use este identificador para citar o enlazar a este item:
http://hdl.handle.net/10261/220014
COMPARTIR / EXPORTAR:
SHARE CORE BASE | |
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL | DATACITE | |
Título: | PreVIous: A Methodology for Prediction of Visual Inference Performance on IoT Devices |
Autor: | Velasco-Montero, Delia CSIC ORCID; Fernández-Berni, J. CSIC ORCID CVN; Carmona-Galán, R. CSIC ORCID ; Rodríguez-Vázquez, Ángel CSIC ORCID | Fecha de publicación: | 2020 | Editor: | Institute of Electrical and Electronics Engineers | Citación: | IEEE Internet of Things Journal (2020) | Resumen: | This paper presents PreVIous, a methodology to predict the performance of Convolutional Neural Networks (CNNs) in terms of throughput and energy consumption on vision-enabled devices for the Internet of Things. CNNs typically constitute a massive computational load for such devices, which are characterized by scarce hardware resources to be shared among multiple concurrent tasks. Therefore, it is critical to select the optimal CNN architecture for a particular hardware platform according to prescribed application requirements. However, the zoo of CNN models is already vast and rapidly growing. To facilitate a suitable selection, we introduce a prediction framework that allows to evaluate the performance of CNNs prior to their actual implementation. The proposed methodology is based on PreVIousNet, a neural network specifically designed to build accurate per-layer performance predictive models. PreVIousNet incorporates the most usual parameters found in state-of-the-art network architectures. The resulting predictive models for inference time and energy have been tested against comprehensive characterizations of seven well-known CNN models running on two different software frameworks and two different embedded platforms. To the best of our knowledge, this is the most extensive study in the literature concerning CNN performance prediction on low-power low-cost devices. The average deviation between predictions and real measurements is remarkably low, ranging from 3% to 10%. This means state-of-the-art modeling accuracy. As an additional asset, the fine-grained a priori analysis provided by PreVIous could also be exploited by neural architecture search engines. | Versión del editor: | https://doi.org/10.1109/JIOT.2020.2981684 | URI: | http://hdl.handle.net/10261/220014 | DOI: | 10.1109/JIOT.2020.2981684 |
Aparece en las colecciones: | (IMSE-CNM) Artículos |
Ficheros en este ítem:
Fichero | Descripción | Tamaño | Formato | |
---|---|---|---|---|
Berni2.pdf | 1,68 MB | Adobe PDF | Visualizar/Abrir |
CORE Recommender
SCOPUSTM
Citations
13
checked on 20-abr-2024
WEB OF SCIENCETM
Citations
9
checked on 29-feb-2024
Page view(s)
178
checked on 24-abr-2024
Download(s)
281
checked on 24-abr-2024
Google ScholarTM
Check
Altmetric
Altmetric
NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.