PAMAP2 dataset preprocessed v0.1.0

Processed PAMAP2 dataset This dataset is based on the [PAMAP2 Dataset for Physical Activity Monitoring](https://archive.ics.uci.edu/ml/datasets/PAMAP2+Physical+Activity+Monitoring). The data is processed with the code in the [mcfly repository] (https://github.com/NLeSC/mcfly/blob/master/tutorial/tutorial_pamap2.py). The preprocessing is similar to that in the paper of Reiss and Stricker, 2012. Windows of length 512 were extracted from within the activity period. Only the variables from the 16g accelerometers on hand, ankle and chest are selected. Only the activities that were performed by all subjects were selected. References A. Reiss and D. Stricker. Introducing a New Benchmarked Dataset for Activity Monitoring. The 16th IEEE International Symposium on Wearable Computers (ISWC), 2012.经过预处理的PAMAP2数据集。本数据集基于[PAMAP2人体活动监测数据集](https://archive.ics.uci.edu/ml/datasets/PAMAP2+Physical+Activity+Monitoring)构建。数据预处理工作依托[mcfly代码仓库](https://github.com/NLeSC/mcfly/blob/master/tutorial/tutorial_pamap2.py)中的代码完成。本次预处理流程与Reiss与Stricker于2012年发表的论文中所采用的方法相近。从活动时段内提取长度为512的时序窗口。仅选取佩戴于手部、脚踝及胸部的16g加速度计对应的特征变量。仅保留所有受试者均完成的活动类别。参考文献：A. Reiss与D. Stricker. 面向活动监测的新型基准数据集. 收录于第16届IEEE可穿戴计算机国际研讨会（ISWC），2012年。

搜集汇总

背景与挑战

背景概述

PAMAP2 dataset preprocessed v0.1.0是一个基于PAMAP2物理活动监测数据集的预处理版本，发布于2017年。该数据集通过提取长度为512的窗口、仅选择手、脚踝和胸部的16g加速度计变量，并筛选所有受试者共同执行的活动进行优化，适用于机器学习任务，如活动识别研究。