PythonPandas：连接包含null的Pandas字符串

Question

我正在尝试连接包含空值的多个字符串列。但结果并没有显示连接的字符串，而只是显示第一列。如何忽略包含null的值？我试图通过替换nan值来处理这个问题，但这没有帮助

import pandas as pd
import numpy as np
df = pd.DataFrame({'Addr1':['a','b','c'], 'Addr2':['', 'x', np.nan], 'Addr3':['202', '201', '203']})
df1 = pd.DataFrame([])
df1['patient address'] = df['Addr1'].astype(str).fillna('',) + " " 
+ df['Addr2'].astype(str).fillna('') + " " 
+ df['Addr3'].astype(str).fillna('') 
print(df1)

预期产出：

      patient address
0              a 202
1              b x 201
2              c 203

Answer 1

import numpy as np
import pandas as pd

df = pd.DataFrame(
    {
        "Addr1": ["a", "b", "c"],
        "Addr2": ["", "x", np.nan],
        "Addr3": ["202", "201", "203"],
    }
)

df1 = pd.DataFrame(
    {
        "patient address": df[["Addr1", "Addr2", "Addr3"]]
        .fillna(" ")
        .agg(" ".join, axis=1)
    }
)

print(df1)

输出：

  patient address
0          a  202
1         b x 201
2         c   203

Answer 2

^{}和groupby+agg如果.join在level=0上，默认情况下stack()会删除NaN：

df1['patient address'] = df.stack().groupby(level=0).agg(' '.join)

---

0     a  202
1    b x 201
2      c 203
dtype: object