CN112543918A - 神经网络切分方法、预测方法及相关装置 - Google Patents

神经网络切分方法、预测方法及相关装置 Download PDF

Info

Publication number
CN112543918A
CN112543918A CN201980013270.7A CN201980013270A CN112543918A CN 112543918 A CN112543918 A CN 112543918A CN 201980013270 A CN201980013270 A CN 201980013270A CN 112543918 A CN112543918 A CN 112543918A
Authority
CN
China
Prior art keywords
sub
network
neural network
subgraph
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201980013270.7A
Other languages
English (en)
Inventor
储洁宇
叶德周
胡雨舟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Publication of CN112543918A publication Critical patent/CN112543918A/zh
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/06—Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5022—Mechanisms to release resources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Neurology (AREA)
  • Image Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种神经网络切分方法、预测方法及相关装置，该神经网络切分方法包括：获得神经网络图，所述神经网络图用于表征一个神经网络；将所述神经网络图进行切分以得到深度子图；所述深度子图包括的多个节点中各节点间通过读写片内缓冲器进行数据交互，所述深度子图用于对由第一输入数据拆分得到的至少两组数据先后进行处理以得到第一输出数据，所述第一输入数据为所述深度子图的输入数据。该方法中，神经网络切分装置切分神经网络图得到一个或多个深度子图，以便于根据该一个或多个深度子图生成一个或多个深度子网络，使用这些深度子网络来执行神经网络的处理任务可以大大减少访问外部存储器的次数。

Description

PCT国内申请，说明书已公开。

Claims (53)

  1. PCT国内申请，权利要求书已公开。
CN201980013270.7A 2019-07-24 2019-12-26 神经网络切分方法、预测方法及相关装置 Pending CN112543918A (zh)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CNPCT/CN2019/097501 2019-07-24
PCT/CN2019/097501 WO2021012215A1 (zh) 2019-07-24 2019-07-24 神经网络切分方法、预测方法及相关装置
PCT/CN2019/128915 WO2021012609A1 (zh) 2019-07-24 2019-12-26 神经网络切分方法、预测方法及相关装置

Publications (1)

Publication Number Publication Date
CN112543918A true CN112543918A (zh) 2021-03-23

Family

ID=74192469

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201980013270.7A Pending CN112543918A (zh) 2019-07-24 2019-12-26 神经网络切分方法、预测方法及相关装置

Country Status (4)

Country Link
US (1) US20220147795A1 (zh)
EP (1) EP3985509A4 (zh)
CN (1) CN112543918A (zh)
WO (2) WO2021012215A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116301903A (zh) * 2023-05-11 2023-06-23 杭州登临瀚海科技有限公司 一种编译器、ai网络编译方法、处理方法、执行系统

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112395282A (zh) * 2019-08-13 2021-02-23 华为技术有限公司 一种图重构方法及装置
US11436830B2 (en) * 2020-03-11 2022-09-06 Bank Of America Corporation Cognitive robotic process automation architecture
US11195080B1 (en) * 2021-03-29 2021-12-07 SambaNova Systems, Inc. Lossless tiling in convolution networks—tiling configuration
US20220388162A1 (en) * 2021-06-08 2022-12-08 Fanuc Corporation Grasp learning using modularized neural networks
US11809521B2 (en) * 2021-06-08 2023-11-07 Fanuc Corporation Network modularization to learn high dimensional robot tasks
CN114648105A (zh) * 2022-02-25 2022-06-21 深圳云天励飞技术股份有限公司 多输出神经网络的切片方法、装置、芯片及存储介质
CN117910523A (zh) * 2022-10-19 2024-04-19 联发科技股份有限公司 将暂存存储器分配给异构设备的方法和系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108292241A (zh) * 2015-10-28 2018-07-17 谷歌有限责任公司 处理计算图
CN108351805A (zh) * 2015-10-28 2018-07-31 谷歌有限责任公司 计算图的基于流的加速器处理
US20180246853A1 (en) * 2017-02-28 2018-08-30 Microsoft Technology Licensing, Llc Hardware node with matrix-vector multiply tiles for neural network processing

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111860812B (zh) * 2016-04-29 2024-03-01 中科寒武纪科技股份有限公司 一种用于执行卷积神经网络训练的装置和方法
US11669727B2 (en) * 2017-01-23 2023-06-06 Nec Corporation Information processing device, neural network design method, and recording medium
CN107832839B (zh) * 2017-10-31 2020-02-14 南京地平线机器人技术有限公司 执行卷积神经网络中的运算的方法和装置
CN107967460B (zh) * 2017-12-08 2020-05-08 重庆广睿达科技有限公司 一种基于深度神经网络的垃圾物焚烧识别方法及系统
CN108388651B (zh) * 2018-02-28 2021-09-28 北京理工大学 一种基于图核和卷积神经网络的文本分类方法
US11698930B2 (en) * 2018-06-21 2023-07-11 Intel Corporation Techniques for determining artificial neural network topologies
CN108876702A (zh) * 2018-06-21 2018-11-23 北京邮电大学 一种加速分布式深度神经网络的训练方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108292241A (zh) * 2015-10-28 2018-07-17 谷歌有限责任公司 处理计算图
CN108351805A (zh) * 2015-10-28 2018-07-31 谷歌有限责任公司 计算图的基于流的加速器处理
US20180246853A1 (en) * 2017-02-28 2018-08-30 Microsoft Technology Licensing, Llc Hardware node with matrix-vector multiply tiles for neural network processing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YU XING ET AL.: "DNNVM: End-to-End Compiler Leveraging Heterogeneous Optimizations on FPGA-Based CNN Accelerators", IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, vol. 39, no. 10, 23 July 2019 (2019-07-23), pages 2668 - 2681, XP011811065, DOI: 10.1109/TCAD.2019.2930577 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116301903A (zh) * 2023-05-11 2023-06-23 杭州登临瀚海科技有限公司 一种编译器、ai网络编译方法、处理方法、执行系统
CN116301903B (zh) * 2023-05-11 2023-08-08 杭州登临瀚海科技有限公司 一种编译器、ai网络编译方法、处理方法、执行系统

Also Published As

Publication number Publication date
WO2021012215A1 (zh) 2021-01-28
US20220147795A1 (en) 2022-05-12
WO2021012609A1 (zh) 2021-01-28
EP3985509A1 (en) 2022-04-20
EP3985509A4 (en) 2022-07-27

Similar Documents

Publication Publication Date Title
CN112543918A (zh) 神经网络切分方法、预测方法及相关装置
CN114186687B (zh) 一种面向神经网络模型计算的中间表示方法和装置
CN110321999B (zh) 神经网络计算图优化方法
CN112199190B (zh) 内存分配方法、装置、存储介质及电子设备
WO2019237811A1 (zh) 一种神经网络的内存分配方法及装置
KR20190055610A (ko) 뉴럴 네트워크 모델들의 공용 연산 그룹을 단일 처리하는 뉴럴 네트워크 시스템, 이를 포함하는 애플리케이션 프로세서 및 뉴럴 네트워크 시스템의 동작방법
US11748599B2 (en) Super-tiling in neural network processing to enable analytics at lower memory speed
CN112711422A (zh) 一种神经网络编译的优化方法及系统
CN112465146B (zh) 一种量子与经典混合云平台以及任务执行方法
CN108776833B (zh) 一种数据处理方法、系统及计算机可读存储介质
CN106709503A (zh) 一种基于密度的大型空间数据聚类算法k‑dbscan
JP2021507345A (ja) 畳み込みニューラル・ネットワークの完全なカーネルを近似するためのスパース・カーネルの融合
CN115423082A (zh) 一种硬件特性相关的深度模型计算图自动优化方法
CN108875914A (zh) 对神经网络数据进行预处理和后处理的方法和装置
CN116204847A (zh) 一种计算图优化方法、装置及设备
CN112819157B (zh) 神经网络训练的方法及装置、智能行驶控制的方法及装置
CN104866297B (zh) 一种优化核函数的方法和装置
KR102326586B1 (ko) 큰 규모 분산 행렬 곱 처리 방법 및 그 장치
CN112200310A (zh) 智能处理器、数据处理方法及存储介质
CN116151363B (zh) 分布式强化学习系统
CN113554157A (zh) 数据处理方法及相关产品
KR102529335B1 (ko) Ai 칩 연계 기반 온디바이스 ai 지원 방법
CN116933841A (zh) 算子融合方法及装置、电子设备、计算机可读介质
CN110955380B (zh) 访存数据生成方法、存储介质、计算机设备和装置
CN113412493A (zh) 基于推理引擎的计算资源分配方法、装置和计算机设备

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination