Skip to Main content Skip to Navigation
Conference papers

MPTorch and MPArchimedes: Open Source Frameworks to Explore Custom Mixed- Precision Operations for DNN Training on Edge Devices

Mariko Tatsumi 1 yuxiang Xie 2 Caroline White 1 Silviu-Ioan Filip 2 Olivier Sentieys 2 Guy Lemieux 1 
1 UBC - University of British Columbia
2 TARAN - Architectures matérielles spécialisées pour l’ère post loi-de-Moore
Inria Rennes – Bretagne Atlantique , IRISA-D3 - ARCHITECTURE
Document type :
Conference papers
Complete list of metadata
https://hal.archives-ouvertes.fr/hal-03494256
Contributor : Olivier Sentieys Connect in order to contact the contributor
Submitted on : Saturday, December 18, 2021 - 7:03:54 PM
Last modification on : Monday, April 4, 2022 - 9:28:32 AM

Identifiers

  • HAL Id : hal-03494256, version 1

Collections

UNIV-RENNES1 | CNRS | INRIA | INSA-RENNES | IRISA | CENTRALESUPELEC | INRIA2 | UR1-MATH-STIC | UR1-UFR-ISTIC | UNIV-RENNES | UR1-MATH-NUM

Citation

Mariko Tatsumi, yuxiang Xie, Caroline White, Silviu-Ioan Filip, Olivier Sentieys, et al.. MPTorch and MPArchimedes: Open Source Frameworks to Explore Custom Mixed- Precision Operations for DNN Training on Edge Devices. ROAD4NN 2021 - 2nd ROAD4NN Workshop: Research Open Automatic Design for Neural Networks, Dec 2021, San Francisco, United States. ⟨hal-03494256⟩

Export

BibTeX TEI DC DCterms EndNote Datacite

Share

Metrics

Record views

70