如何将pandas列分配给其他列，或者如果为nan，则为默认值？

Question

I have

df =

a
1
nan
3

I want some syntax for

df["b"] = df["a"] or 5

to create

a     b
1     1
nan   5
3     3

Does pandas support something like this?

BONUS:

what about different default values per index/group/anything?

Answer 1

import pandas as pd
import numpy as np


df = pd.DataFrame({"a": [1, np.nan, 3]})
df["b"] = df["a"].fillna(5)
print(df)

 ab 0 1.0 1.0 1 NaN 5.0 2 3.0 3.0

digging through the doc gave the standard pandas solution. No need to go through numpy.

Answer 2

You can use np.where

df['b'] = np.where(df['a'].isna(), 5, df['a'])
print(df)

     a    b
0  1.0  1.0
1  NaN  5.0
2  3.0  3.0

Answer 3

您可以applymap()使用applymap()函数甚至factorize()函数。