将嵌套JSON展平到数据帧列中

Question

我有一个带有嵌套json数据字符串的熊猫列。我想将数据展平到多个列中

以下是来自单个单元格的数据：

rent['ques'][9] = "{'Rent': [{'Name': 'Asking', 'Value': 16.07, 'Unit': 'Usd'}], 'Vacancy': {'Name': 'Vacancy', 'Value': 25.34100001, 'Unit': 'Pct'}}"

对于pandas列中的每个单元格，我想解析这个字符串并创建多个列。预期输出如下所示：

当我运行json_normalize(rent['ques'])时，我收到以下错误

 ---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-28-cebc86357f34> in <module>()
----> 1 json_normalize(rentoff['Survey'])

/anaconda3/lib/python3.7/site-packages/pandas/io/json/normalize.py in json_normalize(data, record_path, meta, meta_prefix, record_prefix, errors, sep)
    196     if record_path is None:
    197         if any([[isinstance(x, dict)
--> 198                 for x in compat.itervalues(y)] for y in data]):
    199             # naive normalization, this is idempotent for flat records
    200             # and potentially will inflate the data considerably for

/anaconda3/lib/python3.7/site-packages/pandas/io/json/normalize.py in <listcomp>(.0)
    196     if record_path is None:
    197         if any([[isinstance(x, dict)
--> 198                 for x in compat.itervalues(y)] for y in data]):
    199             # naive normalization, this is idempotent for flat records
    200             # and potentially will inflate the data considerably for

/anaconda3/lib/python3.7/site-packages/pandas/compat/__init__.py in itervalues(obj, **kw)
    210 
    211     def itervalues(obj, **kw):
--> 212         return iter(obj.values(**kw))
    213 
    214     next = next

AttributeError: 'str' object has no attribute 'values'

Answer 1

试试这个：

df['quest'] = df['quest'].str.replace("'", '"')
dfs = []
for i in df['quest']:
    data = json.loads(i)
    dfx = pd.json_normalize(data, record_path=['Rent'], meta=[['Vacancy', 'Name'], ['Vacancy', 'Unit'], ['Vacancy', 'Value']])
    dfs.append(dfx)   

df = pd.concat(dfs).reset_index(drop=['index'])
print(df)


     Name  Value Unit Vacancy.Name Vacancy.Unit Vacancy.Value
0  Asking  16.07  Usd      Vacancy          Pct        25.341
1  Asking  16.07  Usd      Vacancy          Pct        25.341
2  Asking  16.07  Usd      Vacancy          Pct        25.341