Usability and Performance Analysis of Embedded Development Environment for On-device Learning - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2024

Usability and Performance Analysis of Embedded Development Environment for On-device Learning

Évaluation de l'utilisabilité et des performances de différents environnementEmbarqué de développement intégré pour l'apprentissage embarqué.

Résumé

This research empirically examines embedded development tools viable for on-device TinyML implementation. The research evaluates various development tools with various abstraction levels on resource-constrained IoT devices, from basic hardware manipulation to deployment of minimalistic ML training. The analysis encompasses memory usage, energy consumption, and performance metrics during model training and inference and usability of the different solutions. Arduino Framework offers ease of implementation but with increased energy consumption compared to the native option, while RIOT OS exhibits efficient energy consumption despite higher memory utilization with equivalent ease of use. The absence of certain critical functionalities like DVFS directly integrated into the OS highlights limitations for fine hardware control.
Fichier principal
Vignette du fichier
main.pdf (1.11 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY - Paternité

Dates et versions

hal-04505241 , version 1 (14-03-2024)

Licence

Paternité

Identifiants

  • HAL Id : hal-04505241 , version 1

Citer

Enzo Scaffi, Antoine Bonneau, Frédéric Le Mouël, Fabien Mieyeville. Usability and Performance Analysis of Embedded Development Environment for On-device Learning. CITI - CITI Centre of Innovation in Telecommunications and Integration of services; INSA Lyon. 2024. ⟨hal-04505241⟩
3 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More