CPU Utilization Dataset
收藏资源简介:
该数据集是由爱尔兰人工智能中心的科研人员收集的CPU利用率的时序数据,通过在Orange Pi 5硬件上运行psutil库来记录每核心的CPU使用率和UNIX时间戳,持续了大约32天。数据集经过重采样处理以保证每分钟一个样本点,共包含47315个样本点,划分为80%的训练集和20%的测试集。数据集为单变量特征集,利用过去的CPU利用率来预测下一个时间点的CPU利用率。
This dataset is a time-series collection of CPU utilization metrics gathered by researchers at the AI Centre of Ireland. Per-core CPU usage rates and UNIX timestamps were recorded using the psutil library running on Orange Pi 5 hardware, with data collection spanning approximately 32 days. The dataset was resampled to ensure one sample per minute, containing a total of 47,315 samples, which is split into an 80% training set and 20% test set. This is a univariate feature dataset, where the task is to predict the CPU utilization at the next time step based on historical CPU utilization values.




