CN115668222A - 一种神经网络的数据处理方法及装置 - Google Patents
一种神经网络的数据处理方法及装置 Download PDFInfo
- Publication number
- CN115668222A CN115668222A CN202180037755.7A CN202180037755A CN115668222A CN 115668222 A CN115668222 A CN 115668222A CN 202180037755 A CN202180037755 A CN 202180037755A CN 115668222 A CN115668222 A CN 115668222A
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- layer
- layers
- batch
- 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
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Neurology (AREA)
- Image Analysis (AREA)
Abstract
本申请公开了一种神经网络的数据处理方法及装置，涉及人工智能领域。所述方法包括依据输入数据的数据量、运行神经网络的芯片内的内部存储器的第一特征和神经网络中多个层的第二特征动态切分输入数据，为神经网络中的层设置不同的批大小。由于通过为神经网络中的每层设置合理的批大小，在神经网络推理过程中，可充分利用内部存储器存储神经网络的层间数据，从而提高了内部存储器的利用率，以及确保运行神经网络的硬件的计算效率。
Description
PCT国内申请，说明书已公开。
Claims (18)
- PCT国内申请，权利要求书已公开。
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2020/093624 WO2021243489A1 (zh) | 2020-05-30 | 2020-05-30 | 一种神经网络的数据处理方法及装置 |
| CNPCT/CN2020/093624 | 2020-05-30 | ||
| PCT/CN2021/073691 WO2021244045A1 (zh) | 2020-05-30 | 2021-01-26 | 一种神经网络的数据处理方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115668222A true CN115668222A (zh) | 2023-01-31 |
Family
ID=78831421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180037755.7A Pending CN115668222A (zh) | 2020-05-30 | 2021-01-26 | 一种神经网络的数据处理方法及装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115668222A (zh) |
| WO (2) | WO2021243489A1 (zh) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116382880B (zh) * | 2023-06-07 | 2023-08-11 | 成都登临科技有限公司 | 任务执行方法、装置、处理器、电子设备及存储介质 |
| CN117785492B (zh) * | 2024-02-28 | 2024-05-17 | 上海燧原智能科技有限公司 | 一种算子的切分方式确定方法、装置、设备及介质 |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9477925B2 (en) * | 2012-11-20 | 2016-10-25 | Microsoft Technology Licensing, Llc | Deep neural networks training for speech and pattern recognition |
| US10083395B2 (en) * | 2015-05-21 | 2018-09-25 | Google Llc | Batch processing in a neural network processor |
| US10789544B2 (en) * | 2016-04-05 | 2020-09-29 | Google Llc | Batching inputs to a machine learning model |
| JP2018018451A (ja) * | 2016-07-29 | 2018-02-01 | 富士通株式会社 | 機械学習方法、機械学習プログラム及び情報処理装置 |
| CN207440765U (zh) * | 2017-01-04 | 2018-06-01 | 意法半导体股份有限公司 | 片上***和移动计算设备 |
| US20180341852A1 (en) * | 2017-05-24 | 2018-11-29 | International Business Machines Corporation | Balancing memory consumption of multiple graphics processing units in deep learning |
| US11562213B2 (en) * | 2018-04-17 | 2023-01-24 | Intel Corporation | Methods and arrangements to manage memory in cascaded neural networks |
| CN109492754A (zh) * | 2018-11-06 | 2019-03-19 | 深圳市友杰智新科技有限公司 | 一种基于深度神经网络模型压缩和加速方法 |
-
2020
- 2020-05-30 WO PCT/CN2020/093624 patent/WO2021243489A1/zh active Application Filing
-
2021
- 2021-01-26 WO PCT/CN2021/073691 patent/WO2021244045A1/zh active Application Filing
- 2021-01-26 CN CN202180037755.7A patent/CN115668222A/zh active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021243489A1 (zh) | 2021-12-09 |
| WO2021244045A1 (zh) | 2021-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021057713A1 (zh) | 用多核处理器实现神经网络模型拆分方法及相关产品 | |
| CN109102065B (zh) | 一种基于PSoC的卷积神经网络加速器 | |
| US11775430B1 (en) | Memory access for multiple circuit components | |
| JP7366274B2 (ja) | ニューラル・ネットワークのための適応的探索方法および装置 | |
| KR102572757B1 (ko) | 집약성을 개선하기 위한 머신 학습 모델들의 수정 | |
| JP7053775B2 (ja) | ネットワークオンチップによるデータ処理方法及び装置 | |
| US11294599B1 (en) | Registers for restricted memory | |
| CN111465943B (zh) | 一种集成电路和用于神经网络处理的方法 | |
| CN109472361B (zh) | 神经网络优化方法 | |
| CN111105023B (zh) | 数据流重构方法及可重构数据流处理器 | |
| US11853866B2 (en) | Implementation of a neural network in multicore hardware | |
| CN113361695B (zh) | 卷积神经网络加速器 | |
| CN115668222A (zh) | 一种神经网络的数据处理方法及装置 | |
| KR20210148586A (ko) | 스케줄러, 스케줄러의 동작 방법 및 이를 포함한 가속기 시스템 | |
| JP2022137247A (ja) | 複数の入力データセットのための処理 | |
| CN109740619B (zh) | 用于目标识别的神经网络终端运行方法和装置 | |
| US20230143270A1 (en) | Apparatus and method with scheduling | |
| KR20230063791A (ko) | 인공지능 코어, 인공지능 코어 시스템 및 인공지능 코어 시스템의 로드/스토어 방법 | |
| WO2021120036A1 (zh) | 数据处理装置和数据处理方法 | |
| WO2021237755A1 (zh) | 神经网络调度方法及装置 | |
| CN116069393A (zh) | 一种数据处理方法以及相关装置 | |
| WO2020051918A1 (zh) | 神经元电路、芯片、***及其方法、存储介质 | |
| WO2023115529A1 (zh) | 芯片内的数据处理方法及芯片 | |
| EP3985507A1 (en) | Electronic device and method with scheduling | |
| CN210721552U (zh) | 卷积电路 |
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 |