ELM-Research/pretrain-mimic-250-2500

这是一个基于Pretrain MIMIC数据集的随机五折分割数据集，根据患者进行了分层（即训练和测试之间没有患者重叠）。数据集的名称是`pretrain-mimic-250-2500`，其中250表示采样频率，2500表示10秒。数据分割的代码可以在[这里](https://github.com/willxxy/ECG-Bench/blob/main/ecg_bench/create_and_upload_hf.py)找到。如有任何问题或疑问，请联系[ECG-Bench](https://github.com/willxxy)的维护者。

This is a randomly split five fold dataset of the [Pretrain MIMIC dataset](https://www.dropbox.com/scl/fo/ccq5dxmdgg4shf02yjn8c/ANOQ1Hzj4KwHqa1b9r80uzc?rlkey=teysp3v6hg6o9uko2i4zbbjpn&st=exu3i9oo&dl=0) stratified based on patients (i.e., zero patient overlap between training and test). The name of the dataset repository is `pretrain-mimic-250-2500` where 250 refers to the sampling frequency and 2500 denotes 10 seconds. The code to do the splitting is [here](https://github.com/willxxy/ECG-Bench/blob/main/ecg_bench/create_and_upload_hf.py). Any questions or issues, please do not hesitate to reach out to the maintainer of [ECG-Bench](https://github.com/willxxy).