Title:
METHOD AND DEVICE OF COMPUTING LAYOUT SELECTION FOR EFFICIENT DNN INFERENCE
Document Type and Number:
WIPO Patent Application WO/2022/146031
Kind Code:
A1
Abstract:
Embodiments herein provide a method and system for network and hardware aware computing layout selection for efficient Deep Neural Network (DNN) Inference. The method comprises receiving, by the electronic device (100), a DNN model to be executed, wherein the DNN model is associated with a task; dividing the DNN model into a plurality of sub-graphs, wherein each sub-graph is to be processed individually; identifying a computing unit from a plurality of computing units for execution of each sub-graph based on a complexity score; and determining a computing layout from a plurality of computing layouts for each identified computing unit, wherein the sub-graph is executed on the identified computing unit through the determined computing layout.
Inventors:
SINGH BRIRAJ (IN)
UDUPA SHANKARANARAYANA GOPAL AMOGHA (IN)
DWIVEDI ANIKET (IN)
MUDRAGADA BHARAT (IN)
SENAPATI ALLADI ASHOK KUMAR (IN)
KUDRAL SUHAS PARLATHAYA (IN)
ABRAHAM ARUN (IN)
NAIDU PRAVEEN DORESWAMY (IN)
UDUPA SHANKARANARAYANA GOPAL AMOGHA (IN)
DWIVEDI ANIKET (IN)
MUDRAGADA BHARAT (IN)
SENAPATI ALLADI ASHOK KUMAR (IN)
KUDRAL SUHAS PARLATHAYA (IN)
ABRAHAM ARUN (IN)
NAIDU PRAVEEN DORESWAMY (IN)
Application Number:
PCT/KR2021/020175
Publication Date:
July 07, 2022
Filing Date:
December 29, 2021
Export Citation:
Assignee:
SAMSUNG ELECTRONICS CO LTD (KR)
International Classes:
G06N3/08; G06N3/04; G06N5/04
Domestic Patent References:
| WO2016150472A1 | 2016-09-29 |
Foreign References:
| US20200082274A1 | 2020-03-12 | |||
| US20200401445A1 | 2020-12-24 | |||
| US20190205736A1 | 2019-07-04 | |||
| US7114141B1 | 2006-09-26 |
Attorney, Agent or Firm:
KIM, Tae-hun et al. (KR)
Download PDF: