High Gamma Dataset

该数据集包含了14名受试者，每人进行了880次试验的高gamma数据。原始数据使用128个电极记录，但我们仅使用了覆盖运动皮层的44个通道。此外，该数据集被下采样至250赫兹。数据规模为14名受试者，每人880次试验。该任务是对运动皮层活动进行分类。

This dataset consists of high-gamma neural recordings from 14 human subjects, with each subject completing 880 experimental trials. The raw data was initially acquired using 128 electrodes, yet only 44 channels covering the motor cortex are employed here. Additionally, the dataset has been downsampled to 250 Hz. The core task of this dataset is to classify motor cortical neural activities.

搜集汇总

背景与挑战

背景概述

High Gamma Dataset是一个用于脑电图(EEG)解码和可视化研究的公开数据集，包含hdf5格式的脑电数据及预训练模型参数。该数据集与2017年发表在《Human Brain Mapping》期刊上的深度学习研究论文直接相关，并提供了配套的Python加载代码和结果复现示例。