Question

嵌套字典必须转换为pandas df：

columns list = ['city', 'Column_1', 'Column_2',....'Column_20']

{ 'NewYork': {'Column_1': '12146144', 'Column_2': '695',...., 'Column_20': '500'}, 
  'Washington': {'Column_1': '5648', 'Column_2': '864',...., 'Column_20': '734'},
  'Dallas': {'Column_1': '876', 'Column_2': '23456', ...., 'Column_20': '9876'},
  .....
}

尝试了stackoverflow / google中的所有解决方案，但是这种问题在任何地方都不会被问到/解决

list_of_dict = []
for key, value in nested_dict.items():
    for key1, value1 in value.items():
        list_of_dict.append({'city':key, key1:value1})
df = pd.DataFrame(list_of_dict)

预期结果：

 city      Column_1    Column_2  ...  Column_20
New York    12146144    695      ...   500
Washington  5648        864      ...   734
Dallas      876         23456    ...   9876
.
.

但是得到的错误结果是：

city      Column_1    Column_2  ...  Column_20
New York    12146144              
New York                695
New York                        ...     500  
....
Washington  5648        
Washington              864      
Washington                      ...     734
....
Dallas      876         
Dallas                  23456    
Dallas                          ...     9876

Answer 1

您可以使用from_dict

进行检查

pd.DataFrame.from_dict(d,'index').rename_axis('city').reset_index()
Out[119]: 
         city Column_20 Column_2  Column_1
0      Dallas      9876    23456       876
1     NewYork       500      695  12146144
2  Washington       734      864      5648

将深度嵌套的Dictionary转换为pandas数据框架？

1 个答案: