熊猫从词典列表创建数据框

Question

I have a dictionary whose keys are some user IDs and values are lists of dictionaries, take one key-value pair for example:

my_dict['10020'] = [{'type': 'phone', 'count': 3},
                    {'type': 'id_card', 'count': 1},
                    {'type': 'email', 'count': 2}]

Now I would like to create a pandas DataFrame, each row for a key-value pair, columns are the 'type' field within the list of dictionaries above, and values are the 'count' field respectively, like:

    ID    phone    id_card    email
    10020    3           1        2

I have no idea how many potential 'types' are there in the dictionary, so instead of traversing the dictionary and get all 'types', is there a handy way to get the job done?

Answer 1

Data input

d={'10020': [{'type': 'phone', 'count': 3},
                    {'type': 'id_card', 'count': 1},
                    {'type': 'email', 'count': 2}],
 '10021': [{'type': 'phone', 'count': 33},
 {'type': 'id_card', 'count': 11},
{'type': 'email', 'count': 22}]
}

Then we using pd.concate

pd.concat([pd.DataFrame(y).set_index('type').rename(columns={'count':x}).T for x,y in d.items()])


Out[480]: 
type   phone  id_card  email
10020      3        1      2
10021     33       11     22

Answer 2

Consider some data d with variable types:

d = \
{
    "10021": [
        {
            "type": "fax",
            "count": 33
        },
        {
            "type": "email",
            "count": 22
        }
    ],
    "10020": [
        {
            "type": "phone",
            "count": 3
        },
        {
            "type": "id_card",
            "count": 1
        },
        {
            "type": "email",
            "count": 2
        }
    ]
}

Reshape your data as such:

r = [{'id' : k, 'counts' : d[k]} for k in d]    

Now, use json_normalize + pivot :

df = pd.io.json.json_normalize(r, 'counts', 'id').pivot('id', 'type', 'count')
df

type   email   fax  id_card  phone
id                                
10020    2.0   NaN      1.0    3.0
10021   22.0  33.0      NaN    NaN

This should work for any type in your data.