Pandas 系列到 Python 中的数据框

Question

我正在使用 Pandas，但在 Jupyter 中从系列格式化为数据帧时遇到了一些问题。 我有一个具有这种结构的系列

0   {"province": "Paris",
"city": "Paris", "countryCode": "FR", "floor": null, "country":
"France", "route": "RUE MONGE", "extra": null, "coordinates":
[2.35242, 48.84477], "streetNumber": "55", "locationType": null,
"postalCode": "75005"} 
1    {"province": null, "city": "Paris",
"countryCode": "FR", "floor": "CPO_BELI_floor_1482430978123",
"country": "France", "route": "PLACE DU PANTHEON", "extra": null,
"coordinates": [2.345032, 48.845715], "streetNumber": "17",
"locationType": "OUTDOOR", "postalCode": "75005"} 
2    {"province": null, "city": "Paris", "countryCode": "FR", "floor":
"CPO_BELI_floor_1482430978123", "country": "France", "route": "RUE DU
BAC", "extra": null, "coordinates": [2.327753, 48.857124],
"streetNumber": "35", "locationType": "OUTDOOR", "postalCode":
"75007"}

我运行此代码以将其转换为数据帧，但它不会将系列拆分为正确的列：

pd.DataFrame(data['fields.geolocation'], index=data.index)

Answer 1

你很接近，需要将每一行转换为list s：

df = pd.DataFrame(data['fields.geolocation'].values.tolist(), index=data.index)

样品：

a = [{"province": "Paris", "city": "Paris", "countryCode": "FR", "floor": 'null', "country": "France", "route": "RUE MONGE", "extra": 'null', "coordinates": [2.35242, 48.84477], "streetNumber": "55", "locationType": 'null', "postalCode": "75005"} ,
 {"province": 'null', "city": "Paris", "countryCode": "FR", "floor": "CPO_BELI_floor_1482430978123", "country": "France", "route": "PLACE DU PANTHEON", "extra": 'null', "coordinates": [2.345032, 48.845715], "streetNumber": "17", "locationType": "OUTDOOR", "postalCode": "75005"} ,
 {"province": 'null', "city": "Paris", "countryCode": "FR", "floor": "CPO_BELI_floor_1482430978123", "country": "France", "route": "RUE DU BAC", "extra": 'null', "coordinates": [2.327753, 48.857124], "streetNumber": "35", "locationType": "OUTDOOR", "postalCode": "75007"}]

s = pd.Series(a, index=[2,3,5])
print (s)
2    {'province': 'Paris', 'city': 'Paris', 'countr...
3    {'province': 'null', 'city': 'Paris', 'country...
5    {'province': 'null', 'city': 'Paris', 'country...
dtype: object

---

df = pd.DataFrame(s.values.tolist(), index=s.index)
print (df)

    city            coordinates country countryCode extra  \
2  Paris    [2.35242, 48.84477]  France          FR  null   
3  Paris  [2.345032, 48.845715]  France          FR  null   
5  Paris  [2.327753, 48.857124]  France          FR  null   

                          floor locationType postalCode province  \
2                          null         null      75005    Paris   
3  CPO_BELI_floor_1482430978123      OUTDOOR      75005     null   
5  CPO_BELI_floor_1482430978123      OUTDOOR      75007     null   

               route streetNumber  
2          RUE MONGE           55  
3  PLACE DU PANTHEON           17  
5         RUE DU BAC           35  

Answer 2

尝试使用带有axis=1 ( link ) 的pd.concat代替：

这是你的系列：

A = {"province": "Paris", "city": "Paris", "countryCode": "FR", "floor": None, "country": "France", "route": "RUE MONGE", "extra": None, "coordinates": [2.35242, 48.84477], "streetNumber": "55", "locationType": None, "postalCode": "75005"}
B = {"province": None, "city": "Paris", "countryCode": "FR", "floor": "CPO_BELI_floor_1482430978123", "country": "France", "route": "PLACE DU PANTHEON", "extra": None, "coordinates": [2.345032, 48.845715], "streetNumber": "17", "locationType": "OUTDOOR", "postalCode": "75005"}
C = {"province": None, "city": "Paris", "countryCode": "FR", "floor": "CPO_BELI_floor_1482430978123", "country": "France", "route": "RUE DU BAC", "extra": None, "coordinates": [2.327753, 48.857124], "streetNumber": "35", "locationType": "OUTDOOR", "postalCode": "75007"}

A_series = pd.Series(A)
B_series = pd.Series(B)
C_series = pd.Series(C)

这样你就可以创建理想的数据框

df = pd.concat([A_series, B_series, C_series], axis=1)
type(df)
pandas.core.frame.DataFrame

希望这可以帮助。