BECAS
SAAVEDRA Marcos David
congresos y reuniones científicas
Título:
TinyML for Small Microcontrollers
Autor/es:
CÉSAR A. ESTREBOU; MARCOS D. SAAVEDRA; FEDERICO ADRA; MARTÍN FLEMING
Lugar:
La Plata
Reunión:
Conferencia; 10th Conference on Cloud Computing, Big Data & Emerging Topics.; 2022
Institución organizadora:
Instituto de Investigación en Informática LIDI
Resumen:
This paper describes the progress made in the context of a research and development project on machine learning techniques and algorithms applied to small microcontrollers. The beginning of the development of EmbedIA, a machine learning framework for microcontrollers, is presented. The experiments carried out comparing the proposed framework with other similar frameworks such as Tensorflow Lite Micro, uTensor and EloquentTinyML show an important advantage with respect to memory and inference time required by small microcontrollers.
