1  import json
 2  import os
 3  
 4  data_dir = r'E:\Project\Python\MemoTrace\data\聊天记录'
 5  
 6  dev_res = []
 7  train_res = []
 8  
 9  for filepath, dirnames, filenames in os.walk(data_dir):
10      for filename in filenames:
11          if filename.endswith('.json'):
12              print(filename, filepath)
13              filepath_ = os.path.join(filepath, filename)
14              with open(filepath_, 'r', encoding='utf-8') as f:
15                  data = json.load(f)
16              if data:
17                  if filename.endswith('train.json'):
18                      train_res += data
19                  else:
20                      dev_res += data
21  
22  with open('train.json', 'w', encoding='utf-8') as f:
23      json.dump(train_res, f, ensure_ascii=False, indent=4)
24  
25  with open('dev.json', 'w', encoding='utf-8') as f:
26      json.dump(dev_res, f, ensure_ascii=False, indent=4)