arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.05085cs.LGcs.AIcs.PFeess.IV
收藏

移动设备上AI模型和框架的比较和基准测试

Comparison and Benchmarking of AI Models and Frameworks on Mobile Devices

Beijing Academy of Frontier Sciences and Technology(北京前沿科学技术研究院)

Chunjie Luo, Xiwen He, Jianfeng Zhan, Lei Wang, Wanling Gao, Jiahui Dai

摘要:

由于数据量和计算资源的不断增加,深度学习在各个领域取得了许多成功。深度学习在移动设备和嵌入式设备上的应用越来越受到重视,对移动设备和嵌入式设备的AI能力进行基准测试和排名成为亟待解决的问题。考虑到模型的多样性和框架的多样性,我们提出了一个基准套件,AIoTBench,它侧重于评估移动和嵌入式设备的推理能力。AIoTBench涵盖了三种典型的重量级网络:ResNet50,InceptionV3,DenseNet121,以及三种轻量级网络:SqueezeNet,MobileNetV2,MnasNet。每个网络都由三个专为移动和嵌入式设备设计的框架实现:Tensorflow Lite,Caffe2,Pytorch Mobile。为了比较和排名设备的AI能力,我们提出了两个统一的指标作为AI分数:每秒有效图像数(VIPS)和每秒有效浮点数(VOPS)。目前,我们已经使用我们的基准对5款移动设备进行了比较和排名。该列表将在不久之后扩展和更新。

英文摘要:

Due to increasing amounts of data and compute resources, deep learning achieves many successes in various domains. The application of deep learning on the mobile and embedded devices is taken more and more attentions, benchmarking and ranking the AI abilities of mobile and embedded devices becomes an urgent problem to be solved. Considering the model diversity and framework diversity, we propose a benchmark suite, AIoTBench, which focuses on the evaluation of the inference abilities of mobile and embedded devices. AIoTBench covers three typical heavy-weight networks: ResNet50, InceptionV3, DenseNet121, as well as three light-weight networks: SqueezeNet, MobileNetV2, MnasNet. Each network is implemented by three frameworks which are designed for mobile and embedded devices: Tensorflow Lite, Caffe2, Pytorch Mobile. To compare and rank the AI capabilities of the devices, we propose two unified metrics as the AI scores: Valid Images Per Second (VIPS) and Valid FLOPs Per Second (VOPS). Currently, we have compared and ranked 5 mobile devices using our benchmark. This list will be extended and updated soon after.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑