An Efficient FPGA-Based Accelerator for Swin Transformer

现场可编程门阵列 计算 计算机科学 Softmax函数 硬件加速 加速 变压器 规范化(社会学) 推论 并行计算 查阅表格 卷积神经网络 计算科学 计算机工程 高效能源利用 计算机硬件 算法 人工智能 电压 电气工程 工程类 人类学 社会学 程序设计语言
作者
Zhiyang Liu,Pengyu Yin,Zhenhua Ren
出处
期刊:Cornell University - arXiv 被引量:3
标识
DOI:10.48550/arxiv.2308.13922
摘要

Since introduced, Swin Transformer has achieved remarkable results in the field of computer vision, it has sparked the need for dedicated hardware accelerators, specifically catering to edge computing demands. For the advantages of flexibility, low power consumption, FPGAs have been widely employed to accelerate the inference of convolutional neural networks (CNNs) and show potential in Transformer-based models. Unlike CNNs, which mainly involve multiply and accumulate (MAC) operations, Transformer involve non-linear computations such as Layer Normalization (LN), Softmax, and GELU. These nonlinear computations do pose challenges for accelerator design. In this paper, to propose an efficient FPGA-based hardware accelerator for Swin Transformer, we focused on using different strategies to deal with these nonlinear calculations and efficiently handling MAC computations to achieve the best acceleration results. We replaced LN with BN, Given that Batch Normalization (BN) can be fused with linear layers during inference to optimize inference efficiency. The modified Swin-T, Swin-S, and Swin-B respectively achieved Top-1 accuracy rates of 80.7%, 82.7%, and 82.8% in ImageNet. Furthermore, We employed strategies for approximate computation to design hardware-friendly architectures for Softmax and GELU computations. We also designed an efficient Matrix Multiplication Unit to handle all linear computations in Swin Transformer. As a conclude, compared with CPU (AMD Ryzen 5700X), our accelerator achieved 1.76x, 1.66x, and 1.25x speedup and achieved 20.45x, 18.60x, and 14.63x energy efficiency (FPS/power consumption) improvement on Swin-T, Swin-S, and Swin-B models, respectively. Compared to GPU (Nvidia RTX 2080 Ti), we achieved 5.05x, 4.42x, and 3.00x energy efficiency improvement respectively. As far as we know, the accelerator we proposed is the fastest FPGA-based accelerator for Swin Transformer.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
JJXIONG完成签到,获得积分10
1秒前
河马发布了新的文献求助10
2秒前
2秒前
可爱的函函应助肖肖采纳,获得10
3秒前
CodeCraft应助平常的紫蓝采纳,获得10
3秒前
好宝宝完成签到,获得积分10
3秒前
3237924531发布了新的文献求助10
4秒前
hdy331完成签到,获得积分10
4秒前
完美世界应助123采纳,获得10
5秒前
5秒前
怕孤独的忆南完成签到,获得积分10
7秒前
追风完成签到 ,获得积分10
8秒前
8秒前
yao完成签到,获得积分10
8秒前
常芹发布了新的文献求助10
9秒前
ED应助科研路漫漫采纳,获得10
9秒前
10秒前
布鲁克完成签到,获得积分10
11秒前
11秒前
量子星尘发布了新的文献求助10
12秒前
今后应助朴实的绣连采纳,获得30
12秒前
<小天才>发布了新的文献求助10
14秒前
14秒前
15秒前
Smy完成签到 ,获得积分10
15秒前
在水一方应助梁晓雯采纳,获得10
16秒前
Yy123发布了新的文献求助10
16秒前
tao发布了新的文献求助10
16秒前
3237924531完成签到,获得积分10
16秒前
健忘小霜完成签到,获得积分10
17秒前
18秒前
scholar完成签到,获得积分10
19秒前
wei发布了新的文献求助10
19秒前
鳗鱼灵阳完成签到,获得积分20
20秒前
20秒前
21秒前
无情的聋五完成签到 ,获得积分10
21秒前
Owen应助QQiang6采纳,获得10
22秒前
22秒前
SciGPT应助wudizhuzhu233采纳,获得10
22秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
‘Unruly’ Children: Historical Fieldnotes and Learning Morality in a Taiwan Village (New Departures in Anthropology) 400
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
基于可调谐半导体激光吸收光谱技术泄漏气体检测系统的研究 350
Robot-supported joining of reinforcement textiles with one-sided sewing heads 320
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3988838
求助须知:如何正确求助?哪些是违规求助? 3531250
关于积分的说明 11252914
捐赠科研通 3269838
什么是DOI,文献DOI怎么找? 1804820
邀请新用户注册赠送积分活动 881943
科研通“疑难数据库(出版商)”最低求助积分说明 809028