读取记事本里面的内容，并按照要求存放

代码实现

#!/usr/bin/env python
# -*- encoding: utf-8 -*-
'''
@Filename :try.py
@Description :
@Datatime :2021/11/17 09:58:49
@Author :qtxu
@Version :v1.0
@Function :
'''



def _read_data(input_file,input_par_file):
    """Read a BIO data!"""
    """input_file: contains the original words,labels which formatter is word,O,label"""
    """ input_par_file： contains the back-translation words not labels"""
        
    rf = open(input_file,'r')
    rf_par = open(input_par_file, 'r')
    lines = [];words = [];labels = []
    labels_op = []
    par_text = []
    for line in rf:
        # here we dont do "DOCSTART" check
        if len(line.strip()) == 0 :
            l = ' '.join([label for label in labels if len(label) > 0])
            # l_op = ' '.join([label for label in labels_op if len(label) > 0])
            w = ' '.join([word for word in words if len(word) > 0])
            lines.append((l,w))
            words=[]
            labels = []
            labels_op = []
        else:
            word = line.strip().split(' ')[0] # the first str is word in the dataset
            label = line.strip().split(' ')[-1] # the last str is label in the dataset
            # label_op = line.strip().split(' ')[1]
            words.append(word)
            labels.append(label)
            # labels_op.append(label_op)

    for line in rf_par:
        if (len(line.strip()) != 0):
            par_text.append(line.strip())

    rf.close()
    rf_par.close()
    return lines,par_text


if __name__ == "__main__":
    input_file="/home/qtxu/BERT-par/data/15res/train.txt"
    input_par_file = "/home/qtxu/BERT-par/data/15res/15res_train_par_cut.txt"
    lines, par_txt = _read_data(input_file, input_par_file)
    print(lines)
    print(par_txt)

解读

其中，input_file 记事本数据存放格式：

input_par_file 记事本数据存放格式

lines, 数据存放格式

par_txt 数据存放格式