BestOf: an online implementation selector for the training and inference of deep neural networks
Visualitza/
Impacte
Scholar |
Altres documents de l'autoria: Barrachina Mir, Sergio; Castelló, Adrián; Dolz, Manuel F.; Tomás, Andrés E.
Metadades
Mostra el registre complet de l'elementcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/8620
comunitat-uji-handle4:
INVESTIGACIONMetadades
Títol
BestOf: an online implementation selector for the training and inference of deep neural networksData de publicació
2022-05-20Editor
SpringerISSN
0920-8542; 1573-0484Cita bibliogràfica
Barrachina, S., Castelló, A., Dolz, M.F. et al. BestOf: an online implementation selector for the training and inference of deep neural networks. J Supercomput (2022). https://doi.org/10.1007/s11227-022-04577-2Tipus de document
info:eu-repo/semantics/articleVersió
info:eu-repo/semantics/publishedVersionParaules clau / Matèries
Resum
Tuning and optimising the operations executed in deep learning frameworks is a fundamental task in accelerating the processing of deep neural networks (DNNs). However, this optimisation usually requires extensive ... [+]
Tuning and optimising the operations executed in deep learning frameworks is a fundamental task in accelerating the processing of deep neural networks (DNNs). However, this optimisation usually requires extensive manual efforts in order to obtain the best performance for each combination of tensor input size, layer type, and hardware platform. In this work, we present BestOf, a novel online auto-tuner that optimises the training and inference phases of DNNs. BestOf automatically selects at run time, and among the provided alternatives, the best performing implementation in each layer according to gathered profiling data. The evaluation of BestOf is performed on multi-core architectures for different DNNs using PyDTNN, a lightweight library for distributed training and inference. The experimental results reveal that the BestOf auto-tuner delivers the same or higher performance than that achieved using a static selection approach. [-]
Entitat finançadora
Generalitat Valenciana | Ministerio de Ciencia e Innovación
Codi del projecte o subvenció
PID2020-113656RB-C21/C22 | CDEIGENT/2018/014 | FJC2019-039222-I
Drets d'accés
© The Author(s) 2022
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess
Apareix a les col.leccions
- ICC_Articles [430]