当从数据帧中提取单个值时，用什么方法替换`nan`？

I have a dataframe containing NaN values as well as floats

我有一个包含NaN值和浮点数的数据帧

In: import pandas as pd
    import numpy as np
    df = pd.DataFrame([[1,np.nan],[np.nan,4]],columns=['col1','col2'],index=    ['row1','row2'])
    df
Out:
    col1    col2
row1    1.0     NaN
row2    NaN     4.0

To extract a row, or column and fill NaN values, I can use .fillna()

要提取行,列和填充NaN值,我可以使用.fillna()

In: df.loc[:,'col1'].fillna(99.0)
Out: 
    row1     1.0
    row2    99.0
Name: col1, dtype: float64

But .fillna() will not work on a single value. So I have this:

但.fillna()不会对单个值起作用。所以我有这个:

In:
    if np.isnan(df.loc['row2','col1']):
       val = 99.0
    else:
        val = df.loc['row2','col1']
    val

Out:    99.0

This works, but is very clumsy compared to .fillna(). What is the pythonic way?

这有效,但与.fillna()相比非常笨拙。什么是pythonic方式?

1 个解决方案

#1

 df.fillna({'col1':99}, limit=1)

      col1  col2
row1   1.0   NaN
row2  99.0   4.0