数据集获取

git clone https://github.com/qiliang28-dev/Research-on-PVA.git

curl -L -o repo.zip https://github.com/qiliang28-dev/Research-on-PVA/archive/refs/heads/main.zip
unzip repo.zip

源站 README 摘录（使用方式）

Research-on-PVA

Research on patient ventilator asynchrony: a novel respiratory dataset and four non-linear feature extraction methods.
This file introduces the method that how to use the repo.
All the package dependencies were listed in requirements.txt.
The features extracted from raw data have been uploaded for the convenient to skip front 4 steps to quickly get the results.
The novel respiratory dataset was zipped in a file dataset.7z which is of size beyond 20MB, whereas it will nearly occupy 90
MB of memory after releasing it, and the description about it ,including overview.xlsx,was deposited in ./dataset/description/.

Unzip dataset

1. The file ./dataset/raw_data.7z should initially be unzipped to the directory ./raw_data/original_data3/

Data Segmentation

1. Open data_processing.py file, and focus your eyes on the entrance of main function, in which there are two functions
   included, namely uncommented function of start_pre_process() and commented function of start_extract_fea(). The two
   functions can not uncommented at the same time as shown in following code. The function start_pre_process is used to
   preprocess the dataset as you see in the comment, whose argument represents various groups that you want to segment.

    if __name__ == ''__main__'':
        # data segmentation
        start_pre_process(62, 62)
        """
        # extract features automatically
        start_extract_fea(is_normalization=True)
        """
   

2. Different groups of data might need different algorithms to segment.

       def start(self):
           self.calc_real_fs()
           self.update_win_size()
           # self.get_index4(dependency="tidal", back_point=10)
           # self.get_index4(dependency="flow", back_point=0)
           # self.get_index4(dependency="pressure", back_point=3)
           self.get_index()
           self.save_data()
   

   The subfunction above belongs to class SegData. The default method is self.get_index() that can directly split the
   raw data set into breath-by-breath which may contain some improper segmentation.
   The code self.get_index4() is the entrance of another segmenting algorithm. It is up to the argument of dependency
   that can render specific method of segmentation.
3. The resulting file will be appeared in the directory ./original_data3/i

数据加载示例（表格/文本类）

import pandas as pd, glob, os

files = (glob.glob(os.path.join(path, "**", "*.csv"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tsv"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.xlsx"), recursive=True))
print("数据文件:", files)
df = pd.read_csv(files[0])
print(df.shape); print(df.columns.tolist()); print(df.head(3))

完整仓库：github.com/qiliang28-dev/Research-on-PVA