主页 / 问题 / 1090198
Accepted
Денис Астаховский
Asked:2020-03-03 21:23:40 +0000 UTC2020-03-03 21:23:40 +0000 UTC

为什么替换字符时会得到 NaN

  • 772
import pandas as pd
data = {'kol_click':[1, 8, 4, 2, 1, '', 18, '', 3, 10]}
df1 = pd.DataFrame(data)
df1['kol_click_1'] = df1['kol_click'].str.replace('', '0')
print(df1.head(10))

  kol_click kol_click_1
0         1         NaN
1         8         NaN
2         4         NaN
3         2         NaN
4         1         NaN
5                     0
6        18         NaN
7                     0
8         3         NaN
9        10         NaN

为什么数字变为 NaN ？如何用空值替换空字符串？

python

1 个回答

  1. Best Answer

    该问题是由在同一列中混合整数和字符串引起的。Pandas 将此类列的类型视为object：

    In [17]: df1.dtypes
Out[17]:
kol_click    object
dtype: object

    但是不可能像使用常规字符串列那样舒适地使用这样的列：

    In [25]: df1['kol_click'].str[:10]
Out[25]:
0    NaN
1    NaN
2    NaN
3    NaN
4    NaN
5
6    NaN
7
8    NaN
9    NaN
Name: kol_click, dtype: object

In [26]: df1['kol_click'].astype(str).str[:10]
Out[26]:
0     1
1     8
2     4
3     2
4     1
5
6    18
7
8     3
9    10
Name: kol_click, dtype: object

    解决方案- 试试这个：

    In [22]: df1['kol_click_1'] = pd.to_numeric(df1['kol_click'], errors='coerce').fillna(0)

In [23]: df1
Out[23]:
  kol_click  kol_click_1
0         1          1.0
1         8          8.0
2         4          4.0
3         2          2.0
4         1          1.0
5                    0.0
6        18         18.0
7                    0.0
8         3          3.0
9        10         10.0

In [24]: df1.dtypes
Out[24]:
kol_click       object
kol_click_1    float64
dtype: object
    • 1

相关问题