MLPerf Tiny Benchmark Suite
收藏arXiv2025-09-30 收录
下载链接:
https://mlcommons.org/en/mlperf-tiny/
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是MLPerf Tiny基准测试套件的一部分,涵盖了四个网络,用于评估边缘设备上深度神经网络(DNN)的性能。其中包括针对CIFAR10的ResNet V1模型、针对视觉唤醒词的MobileNetV1模型、针对关键词识别的DS-CNN模型,以及针对DCASE2020的的全连接自动编码器。这些模型的任务涵盖了图像分类和目标检测。
This dataset is part of the MLPerf Tiny benchmark suite, which includes four networks for evaluating the performance of deep neural networks (DNNs) on edge devices. Specifically, it covers the ResNet V1 model for CIFAR-10, the MobileNetV1 model for visual wake words, the DS-CNN model for keyword spotting, and the fully connected autoencoder for DCASE 2020. The tasks of these models span image classification and object detection.
提供机构:
MLPerf



