Cargando…

HLS4ML - Testing Infrastructure

Testing is a process to execute a software or program and find all the errors and bugs in software/program which do not meet the requirements or have inconsistency during the executing of the program. With more uses of machine learning and deep neural networks in the particle physics sector, High-Le...

Descripción completa

Detalles Bibliográficos
Autor principal: Nuntaviriyakul, Sarun
Lenguaje:eng
Publicado: 2020
Materias:
Acceso en línea:http://cds.cern.ch/record/2743530
_version_ 1780968560107978752
author Nuntaviriyakul, Sarun
author_facet Nuntaviriyakul, Sarun
author_sort Nuntaviriyakul, Sarun
collection CERN
description Testing is a process to execute a software or program and find all the errors and bugs in software/program which do not meet the requirements or have inconsistency during the executing of the program. With more uses of machine learning and deep neural networks in the particle physics sector, High-Level Synthesis languages for FPGAs called hls4ml are used. With more layers and frameworks being supported, it is crucial to maintain the consistency and the functionality of the software when there is new change added. To ensure that the conversion process of deep neural networks is being executed correctly we create a testing infrastructure and testing pipeline for the conversion of deep neural networks using hls4ml.
id cern-2743530
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2020
record_format invenio
spelling cern-27435302020-11-04T20:05:25Zhttp://cds.cern.ch/record/2743530engNuntaviriyakul, SarunHLS4ML - Testing InfrastructurePhysics in GeneralTesting is a process to execute a software or program and find all the errors and bugs in software/program which do not meet the requirements or have inconsistency during the executing of the program. With more uses of machine learning and deep neural networks in the particle physics sector, High-Level Synthesis languages for FPGAs called hls4ml are used. With more layers and frameworks being supported, it is crucial to maintain the consistency and the functionality of the software when there is new change added. To ensure that the conversion process of deep neural networks is being executed correctly we create a testing infrastructure and testing pipeline for the conversion of deep neural networks using hls4ml.CERN-STUDENTS-Note-2020-033oai:cds.cern.ch:27435302020-11-04
spellingShingle Physics in General
Nuntaviriyakul, Sarun
HLS4ML - Testing Infrastructure
title HLS4ML - Testing Infrastructure
title_full HLS4ML - Testing Infrastructure
title_fullStr HLS4ML - Testing Infrastructure
title_full_unstemmed HLS4ML - Testing Infrastructure
title_short HLS4ML - Testing Infrastructure
title_sort hls4ml - testing infrastructure
topic Physics in General
url http://cds.cern.ch/record/2743530
work_keys_str_mv AT nuntaviriyakulsarun hls4mltestinginfrastructure