Question

以下是我的数据框的输出：

              0                                  1  
0  {"time": "2016-03-28T23:23:12Z"      "target": "Raffi-Antilian"}  
1  {"time": "2016-03-28T23:23:12Z"      "target": "Caroline-Kaiser"}

如何将单个记录从类型字典转换为普通数据帧记录，列名是字典键，记录值是字典值？我想要的输出应该是：

              Time                Target  
0  2016-03-28T23:23:12Z      Raffi-Antilian   
1  2016-03-28T23:23:12Z      Caroline-Kaiser

我有大约2000条记录，感谢任何帮助/指导。

Answer 1

import json
data = []
with open('filename', 'r') as f:
    for line in f:
        data.append(json.loads(line))
pd.DataFrame(data)

给出

Out[49]: 
            target                  time
0   Raffi-Antilian  2016-03-28T23:23:12Z
1  Caroline-Kaiser  2016-03-28T23:23:12Z

Answer 2

如果文件不 sep=';'，则read_csv ;可以Help Center，因此所有数据都在一个Series中。然后按string将dictionary转换为ast.literal_eval，最后使用pd.DataFrame：

import pandas as pd
import ast
import io

temp=u"""{"time": "2016-03-28T23:23:12Z","target": "Raffi-Antilian"}  
{"time": "2016-03-28T23:23:12Z","target": "Caroline-Kaiser"}"""
#after testing replace io.StringIO(temp) to filename
s = pd.read_csv(io.StringIO(temp), index_col=None, header=None, sep=';', squeeze=True)
print (s)
0    {"time": "2016-03-28T23:23:12Z","target": "Raf...
1    {"time": "2016-03-28T23:23:12Z","target": "Car...
Name: 0, dtype: object

L = s.apply(lambda x: ast.literal_eval(x)).tolist()
print (L)
[{'time': '2016-03-28T23:23:12Z', 'target': 'Raffi-Antilian'}, 
 {'time': '2016-03-28T23:23:12Z', 'target': 'Caroline-Kaiser'}]

print (pd.DataFrame(L))
            target                  time
0   Raffi-Antilian  2016-03-28T23:23:12Z
1  Caroline-Kaiser  2016-03-28T23:23:12Z

编辑：

另一个单行解决方案：

import pandas as pd
import json

print (pd.DataFrame([json.loads(line.strip()) for line in open('file.txt')]))

            target                  time
0   Raffi-Antilian  2016-03-28T23:23:12Z
1  Caroline-Kaiser  2016-03-28T23:23:12Z

从pandas dataframe中删除字典引用

2 个答案: