CN115374935A - 一种神经网络的剪枝方法 - Google Patents
一种神经网络的剪枝方法 Download PDFInfo
- Publication number
- CN115374935A CN115374935A CN202211122342.1A CN202211122342A CN115374935A CN 115374935 A CN115374935 A CN 115374935A CN 202211122342 A CN202211122342 A CN 202211122342A CN 115374935 A CN115374935 A CN 115374935A
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- training
- channel
- accelerator
- 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.)
- Granted
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
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Neurology (AREA)
- Complex Calculations (AREA)
Abstract
Description
Claims (7)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211122342.1A CN115374935B (zh) | 2022-09-15 | 2022-09-15 | 一种神经网络的剪枝方法 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211122342.1A CN115374935B (zh) | 2022-09-15 | 2022-09-15 | 一种神经网络的剪枝方法 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115374935A true CN115374935A (zh) | 2022-11-22 |
CN115374935B CN115374935B (zh) | 2023-08-11 |
Family
ID=84072412
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211122342.1A Active CN115374935B (zh) | 2022-09-15 | 2022-09-15 | 一种神经网络的剪枝方法 |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115374935B (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024150875A1 (ko) * | 2023-01-11 | 2024-07-18 | 주식회사 사피온코리아 | 시스톨릭 어레이와 메모리 간의 데이터 전달을 위한 방법 및 장치 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180336468A1 (en) * | 2017-05-16 | 2018-11-22 | Nec Laboratories America, Inc. | Pruning filters for efficient convolutional neural networks for image recognition in surveillance applications |
CN110796251A (zh) * | 2019-10-28 | 2020-02-14 | 天津大学 | 基于卷积神经网络的图像压缩优化方法 |
CN112183744A (zh) * | 2020-09-25 | 2021-01-05 | 中国科学院计算技术研究所 | 一种神经网络剪枝方法及装置 |
US20210097375A1 (en) * | 2019-09-27 | 2021-04-01 | Amazon Technologies, Inc. | Transposed convolution using systolic array |
US20220012593A1 (en) * | 2019-07-08 | 2022-01-13 | Zhejiang University | Neural network accelerator and neural network acceleration method based on structured pruning and low-bit quantization |
CN114662689A (zh) * | 2022-03-31 | 2022-06-24 | 重庆大学 | 一种神经网络的剪枝方法、装置、设备及介质 |
WO2022141754A1 (zh) * | 2020-12-31 | 2022-07-07 | 之江实验室 | 一种卷积神经网络通用压缩架构的自动剪枝方法及平台 |
CN114925823A (zh) * | 2022-05-12 | 2022-08-19 | 南京航空航天大学 | 一种卷积神经网络压缩方法及边缘侧fpga加速器 |
-
2022
- 2022-09-15 CN CN202211122342.1A patent/CN115374935B/zh active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180336468A1 (en) * | 2017-05-16 | 2018-11-22 | Nec Laboratories America, Inc. | Pruning filters for efficient convolutional neural networks for image recognition in surveillance applications |
US20220012593A1 (en) * | 2019-07-08 | 2022-01-13 | Zhejiang University | Neural network accelerator and neural network acceleration method based on structured pruning and low-bit quantization |
US20210097375A1 (en) * | 2019-09-27 | 2021-04-01 | Amazon Technologies, Inc. | Transposed convolution using systolic array |
CN110796251A (zh) * | 2019-10-28 | 2020-02-14 | 天津大学 | 基于卷积神经网络的图像压缩优化方法 |
CN112183744A (zh) * | 2020-09-25 | 2021-01-05 | 中国科学院计算技术研究所 | 一种神经网络剪枝方法及装置 |
WO2022141754A1 (zh) * | 2020-12-31 | 2022-07-07 | 之江实验室 | 一种卷积神经网络通用压缩架构的自动剪枝方法及平台 |
CN114662689A (zh) * | 2022-03-31 | 2022-06-24 | 重庆大学 | 一种神经网络的剪枝方法、装置、设备及介质 |
CN114925823A (zh) * | 2022-05-12 | 2022-08-19 | 南京航空航天大学 | 一种卷积神经网络压缩方法及边缘侧fpga加速器 |
Non-Patent Citations (2)
Title |
---|
FENG SHI 等: "Sparse Winograd Convolutional neural networks on small-scale systolic arrays", ARXIV:1810.01973V1, pages 1 - 7 * |
H. T. KUNG: "Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimization", MULTIPLIERARXIV: 1811.04770V1, pages 1 - 13 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024150875A1 (ko) * | 2023-01-11 | 2024-07-18 | 주식회사 사피온코리아 | 시스톨릭 어레이와 메모리 간의 데이터 전달을 위한 방법 및 장치 |
Also Published As
Publication number | Publication date |
---|---|
CN115374935B (zh) | 2023-08-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111242289B (zh) | 一种规模可扩展的卷积神经网络加速系统与方法 | |
CN111445012B (zh) | 一种基于fpga的分组卷积硬件加速器及其方法 | |
CN110390385B (zh) | 一种基于bnrp的可配置并行通用卷积神经网络加速器 | |
CN110348574B (zh) | 一种基于zynq的通用卷积神经网络加速结构及设计方法 | |
CN109409510B (zh) | 神经元电路、芯片、系统及其方法、存储介质 | |
CN111898733B (zh) | 一种深度可分离卷积神经网络加速器架构 | |
CN112149811A (zh) | 调度感知张量分发模块 | |
CN111079923B (zh) | 适用于边缘计算平台的Spark卷积神经网络系统及其电路 | |
CN113220630B (zh) | 一种硬件加速器的可重构阵列优化方法及自动调优方法 | |
CN110222818A (zh) | 一种用于卷积神经网络数据存储的多bank行列交织读写方法 | |
CN108304925B (zh) | 一种池化计算装置及方法 | |
CN113240101B (zh) | 卷积神经网络软硬件协同加速的异构SoC实现方法 | |
CN111768458A (zh) | 一种基于卷积神经网络的稀疏图像处理方法 | |
CN109740619B (zh) | 用于目标识别的神经网络终端运行方法和装置 | |
CN111210019A (zh) | 一种基于软硬件协同加速的神经网络推断方法 | |
CN115374935A (zh) | 一种神经网络的剪枝方法 | |
JP7332722B2 (ja) | データ処理方法、装置、記憶媒体及び電子機器 | |
CN108520297A (zh) | 可编程深度神经网络处理器 | |
Zhang et al. | Hardware-software codesign of weight reshaping and systolic array multiplexing for efficient CNNs | |
CN113780529B (zh) | 一种面向fpga的稀疏卷积神经网络多级存储计算系统 | |
CN112001492A (zh) | 关于二值权重DenseNet模型的混合流水式加速架构及加速方法 | |
CN114239816B (zh) | 一种卷积神经网络-图卷积神经网络可重构硬件加速架构 | |
CN110728303A (zh) | 基于卷积神经网络数据复杂度的动态自适应计算阵列 | |
Zhu et al. | Optimization of convolutional neural network hardware structure based on FPGA | |
CN116882467B (zh) | 面向边缘端的多模式可配置的神经网络加速器电路结构 |
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 | ||
CB03 | Change of inventor or designer information |
Inventor after: Wang Peng Inventor after: Pu Xingquan Inventor after: Wang Chengliang Inventor after: Wu Hao Inventor after: Yang Chan Inventor after: Huang Zhetong Inventor after: Ren Ao Inventor before: Pu Xingquan Inventor before: Wang Chengliang Inventor before: Wang Peng Inventor before: Wu Hao Inventor before: Yang Chan Inventor before: Huang Zhetong Inventor before: Ren Ao |
|
CB03 | Change of inventor or designer information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |