替换python中的列值

这是一个很容易回答的问题：

我有一个数据框，看起来像这样：

import pandas as pd
names_raw = {
    'device_id': [ '1d28d33a-c98e-4986-a7bb-5881d222c9a8','54322099-e76d-4986-afd2-0861e2113a16','ec3a9f9d-8e4d-4986-bea8-c17c361366e9','cc8e247d-4e2e-4986-b783-e516d03a358c','ca2d8769-ccf5-4986-8aed-741ca68e94cd','12178e22-6d64-4986-966a-374326fdaf3d','50ba7a2e-a1aa-4986-86a7-08e0605dc702','f427c8e9-65d4-46de-b986-8f8e79242842','cee68e2b-135f-45b0-be4b-7c23009866ba','e785988e-2693-47ad-9899-0049860ccaa7','a1986866-13f8-4dbe-b661-8c9f78eac745','a9998ecd-9fe9-4932-870d-29c6b5df1214','9b88e362-b06d-4317-96f5-f266c986a8d6','a04498ef-fd7c-4aa4-bffc-9158ccbad3a1'],
    'pod_id': ['B00001','B00011','B00013','B00016','B00021','B00023','B00024','B00026','B00027','B00028','B00030','B00032','B00034','B00039'],
    'native_id': ['zim_pod_0001','zim_pod_0002',    'zim_pod_0003', 'zim_pod_0004', 'zim_pod_0005', 'zim_pod_0006', 'zim_pod_0007', 'zim_pod_0008', 'zim_pod_0009', 'zim_pod_0010', 'zim_pod_0011', 'zim_pod_0012', 'zim_pod_0013','zim_pod_0014']
    }
names = pd.DataFrame(names_raw, columns = ['device_id', 'pod_id', 'native_id'])

>>> df
                          device_id      day  month  year  rain
0  1d28d33a-c98e-4986-a7bb-5881d222c9a8   31     12  2016   0.0
1  54322099-e76d-4986-afd2-0861e2113a16   31     12  2016   0.0
2  ec3a9f9d-8e4d-4986-bea8-c17c361366e9   31     12  2016   0.0
3  cc8e247d-4e2e-4986-b783-e516d03a358c   31     12  2016   1.2
4  ca2d8769-ccf5-4986-8aed-741ca68e94cd   31     12  2016   2.2
5  12178e22-6d64-4986-966a-374326fdaf3d   31     12  2016   0.2
6  9b88e362-b06d-4317-96f5-f266c986a8d6   31     12  2016   0.0

>>> df
                           native_id      day  month  year  rain
0                          zim_pod_0001   31     12  2016   0.0
1                          zim_pod_0002   31     12  2016   0.0
2                          zim_pod_0003   31     12  2016   0.0

另一个数据帧如下所示：

import pandas as pd
names_raw = {
    'device_id': [ '1d28d33a-c98e-4986-a7bb-5881d222c9a8','54322099-e76d-4986-afd2-0861e2113a16','ec3a9f9d-8e4d-4986-bea8-c17c361366e9','cc8e247d-4e2e-4986-b783-e516d03a358c','ca2d8769-ccf5-4986-8aed-741ca68e94cd','12178e22-6d64-4986-966a-374326fdaf3d','50ba7a2e-a1aa-4986-86a7-08e0605dc702','f427c8e9-65d4-46de-b986-8f8e79242842','cee68e2b-135f-45b0-be4b-7c23009866ba','e785988e-2693-47ad-9899-0049860ccaa7','a1986866-13f8-4dbe-b661-8c9f78eac745','a9998ecd-9fe9-4932-870d-29c6b5df1214','9b88e362-b06d-4317-96f5-f266c986a8d6','a04498ef-fd7c-4aa4-bffc-9158ccbad3a1'],
    'pod_id': ['B00001','B00011','B00013','B00016','B00021','B00023','B00024','B00026','B00027','B00028','B00030','B00032','B00034','B00039'],
    'native_id': ['zim_pod_0001','zim_pod_0002',    'zim_pod_0003', 'zim_pod_0004', 'zim_pod_0005', 'zim_pod_0006', 'zim_pod_0007', 'zim_pod_0008', 'zim_pod_0009', 'zim_pod_0010', 'zim_pod_0011', 'zim_pod_0012', 'zim_pod_0013','zim_pod_0014']
    }
names = pd.DataFrame(names_raw, columns = ['device_id', 'pod_id', 'native_id'])

>>> df
                          device_id      day  month  year  rain
0  1d28d33a-c98e-4986-a7bb-5881d222c9a8   31     12  2016   0.0
1  54322099-e76d-4986-afd2-0861e2113a16   31     12  2016   0.0
2  ec3a9f9d-8e4d-4986-bea8-c17c361366e9   31     12  2016   0.0
3  cc8e247d-4e2e-4986-b783-e516d03a358c   31     12  2016   1.2
4  ca2d8769-ccf5-4986-8aed-741ca68e94cd   31     12  2016   2.2
5  12178e22-6d64-4986-966a-374326fdaf3d   31     12  2016   0.2
6  9b88e362-b06d-4317-96f5-f266c986a8d6   31     12  2016   0.0

>>> df
                           native_id      day  month  year  rain
0                          zim_pod_0001   31     12  2016   0.0
1                          zim_pod_0002   31     12  2016   0.0
2                          zim_pod_0003   31     12  2016   0.0

我想用

native\u id

列替换

device\u id

列。如何使用最少的代码行来实现这一点

最终的数据帧应如下所示：

import pandas as pd
names_raw = {
    'device_id': [ '1d28d33a-c98e-4986-a7bb-5881d222c9a8','54322099-e76d-4986-afd2-0861e2113a16','ec3a9f9d-8e4d-4986-bea8-c17c361366e9','cc8e247d-4e2e-4986-b783-e516d03a358c','ca2d8769-ccf5-4986-8aed-741ca68e94cd','12178e22-6d64-4986-966a-374326fdaf3d','50ba7a2e-a1aa-4986-86a7-08e0605dc702','f427c8e9-65d4-46de-b986-8f8e79242842','cee68e2b-135f-45b0-be4b-7c23009866ba','e785988e-2693-47ad-9899-0049860ccaa7','a1986866-13f8-4dbe-b661-8c9f78eac745','a9998ecd-9fe9-4932-870d-29c6b5df1214','9b88e362-b06d-4317-96f5-f266c986a8d6','a04498ef-fd7c-4aa4-bffc-9158ccbad3a1'],
    'pod_id': ['B00001','B00011','B00013','B00016','B00021','B00023','B00024','B00026','B00027','B00028','B00030','B00032','B00034','B00039'],
    'native_id': ['zim_pod_0001','zim_pod_0002',    'zim_pod_0003', 'zim_pod_0004', 'zim_pod_0005', 'zim_pod_0006', 'zim_pod_0007', 'zim_pod_0008', 'zim_pod_0009', 'zim_pod_0010', 'zim_pod_0011', 'zim_pod_0012', 'zim_pod_0013','zim_pod_0014']
    }
names = pd.DataFrame(names_raw, columns = ['device_id', 'pod_id', 'native_id'])

>>> df
                          device_id      day  month  year  rain
0  1d28d33a-c98e-4986-a7bb-5881d222c9a8   31     12  2016   0.0
1  54322099-e76d-4986-afd2-0861e2113a16   31     12  2016   0.0
2  ec3a9f9d-8e4d-4986-bea8-c17c361366e9   31     12  2016   0.0
3  cc8e247d-4e2e-4986-b783-e516d03a358c   31     12  2016   1.2
4  ca2d8769-ccf5-4986-8aed-741ca68e94cd   31     12  2016   2.2
5  12178e22-6d64-4986-966a-374326fdaf3d   31     12  2016   0.2
6  9b88e362-b06d-4317-96f5-f266c986a8d6   31     12  2016   0.0

>>> df
                           native_id      day  month  year  rain
0                          zim_pod_0001   31     12  2016   0.0
1                          zim_pod_0002   31     12  2016   0.0
2                          zim_pod_0003   31     12  2016   0.0

等等等等。

试试这个：

df['native_id'] = df.device_id.map(names.set_index('device_id')['native_id'])

或者如果您不想在

df

df中保留

device\u id

列：

In [210]: df['native_id'] = df.pop('device_id').map(names.set_index('device_id')['native_id'])

In [211]: df
Out[211]:
   day  month  year  rain     native_id
0   31     12  2016   0.0  zim_pod_0001
1   31     12  2016   0.0  zim_pod_0002
2   31     12  2016   0.0  zim_pod_0003
3   31     12  2016   1.2  zim_pod_0004
4   31     12  2016   2.2  zim_pod_0005
5   31     12  2016   0.2  zim_pod_0006
6   31     12  2016   0.0  zim_pod_0013

使用Pandas内置的

merge（）

方法。它本质上是一个连接，使用起来非常简单。指定device_id作为连接键，然后选择所需的列，如下所示：

df2 = pd.merge(df,names,on="device_id")[["native_id","day","month","year","rain"]]

结果:

      native_id  day  month  year  rain
0  zim_pod_0001   31     12  2016   0.0
1  zim_pod_0002   31     12  2016   0.0
2  zim_pod_0003   31     12  2016   0.0
3  zim_pod_0004   31     12  2016   1.2
4  zim_pod_0005   31     12  2016   2.2
5  zim_pod_0006   31     12  2016   0.2
6  zim_pod_0013   31     12  2016   0.0

---

@JAG2024，很高兴我能帮忙：）