python - Pandas:用组的模式填充 na

Question

我有一个 df多列。

df = pd.DataFrame({'Store':['M1','M2','M3','M1','M1','M2','M2','M3','M3'],
                   'Category':['A','A','A','B','B','B','C','C','C'],
                   'Price_Category':[np.nan,X,np.nan,np.nan,Y,Y,Z,np.nan,Z]})

我如何填写NaN在 Price_Category与模式，仅基于Category ?
我尝试使用:

df['Price_Category'] = df.groupby('Category')['Price_Category'].apply(lambda x: x.fillna(x.mode()[0]))

但我收到此错误:KeyError: 0我试过的其他方法，开始填NaN名称来自 Store .
谢谢你的帮助!

Answer 1

使用 Series.iat 对于 Series 的第一个值按职位:

f = lambda x: x.fillna(x.mode().iat[0])
df['Price_Category'] = df.groupby('Category')['Price_Category'].apply(f)

print (df)
  Store Category Price_Category
0    M1        A              X
1    M2        A              X
2    M3        A              X
3    M1        B              Y
4    M1        B              Y
5    M2        B              Y
6    M2        C              Z
7    M3        C              Z
8    M3        C              Z

如果错误:

IndexError: index 0 is out of bounds for axis 0 with size 0

这意味着只有 NaN 一组或多组(已更改数据以引发错误):

df = pd.DataFrame({'Store':['M1','M2','M3','M1','M1','M2','M2','M3','M3'],
                   'Category':['A','A','A','B','B','B','C','C','C'],
                  'Price_Category':[np.nan,'X',np.nan,np.nan,'Y','Y',np.nan,np.nan,np.nan]})

#test if at least one non NaN else return same values (NaNs)
f = lambda x: x.fillna(x.mode().iat[0]) if x.notna().any() else x
df['Price_Category'] = df.groupby('Category')['Price_Category'].apply(f)

print (df)
  Store Category Price_Category
0    M1        A              X
1    M2        A              X
2    M3        A              X
3    M1        B              Y
4    M1        B              Y
5    M2        B              Y
6    M2        C            NaN
7    M3        C            NaN
8    M3        C            NaN