Geodesic
Performance analysis methodology of deep neural networks inference on the example of an image classification problem
Numerical methods and programming, Tome 25 (2024) no. 2, pp. 127-141.

Voir la notice de l'article provenant de la source Math-Net.Ru

Deploying of deep neural networks requires inference performance analysis on the target hardware. Performance results are aimed to be used as motivation to evaluate a decision for deployment, find the best performing hardware and software configurations, decide is there's a need for optimization of DL model and DL inference software. The paper describes a technique for analyzing and comparing inference performance using an example of image classification problem: converting a trained model to the formats of different frameworks, quality analysis, determining optimal inference execution parameters, model optimization and quality reanalysis, analyzing and comparing inference performance for the considered frameworks. Deep Learning Inference Benchmark Tool is aimed to support the performance analysis cycle. The technique is implemented on the example of the MobileNetV2 model.
Keywords: deep learning, neural networks, inference, performance, MobileNetV2, Deep Learning Inference Benchmark. 
@article{VMP_2024_25_2_a1,
     author = {M. R. Alibekov and N. E. Berezina and E. P. Vasiliev and I. B. Vikhrev and Yu. D. Kamelina and V. D. Kustikova and Z. A. Maslova and I. S. Mukhin and A. K. Sidorova and V. N. Suchkov},
     title = {Performance analysis methodology of deep neural networks inference on the example of an image classification problem},
     journal = {Numerical methods and programming},
     pages = {127--141},
     publisher = {mathdoc},
     volume = {25},
     number = {2},
     year = {2024},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a1/}
}
TY  - JOUR
AU  - M. R. Alibekov
AU  - N. E. Berezina
AU  - E. P. Vasiliev
AU  - I. B. Vikhrev
AU  - Yu. D. Kamelina
AU  - V. D. Kustikova
AU  - Z. A. Maslova
AU  - I. S. Mukhin
AU  - A. K. Sidorova
AU  - V. N. Suchkov
TI  - Performance analysis methodology of deep neural networks inference on the example of an image classification problem
JO  - Numerical methods and programming
PY  - 2024
SP  - 127
EP  - 141
VL  - 25
IS  - 2
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a1/
LA  - ru
ID  - VMP_2024_25_2_a1
ER  - 
%0 Journal Article
%A M. R. Alibekov
%A N. E. Berezina
%A E. P. Vasiliev
%A I. B. Vikhrev
%A Yu. D. Kamelina
%A V. D. Kustikova
%A Z. A. Maslova
%A I. S. Mukhin
%A A. K. Sidorova
%A V. N. Suchkov
%T Performance analysis methodology of deep neural networks inference on the example of an image classification problem
%J Numerical methods and programming
%D 2024
%P 127-141
%V 25
%N 2
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a1/
%G ru
%F VMP_2024_25_2_a1
M. R. Alibekov; N. E. Berezina; E. P. Vasiliev; I. B. Vikhrev; Yu. D. Kamelina; V. D. Kustikova; Z. A. Maslova; I. S. Mukhin; A. K. Sidorova; V. N. Suchkov. Performance analysis methodology of deep neural networks inference on the example of an image classification problem. Numerical methods and programming, Tome 25 (2024) no. 2, pp. 127-141. http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a1/