Pandas Apply Exception

Question

I am trying to build a function to give me the most common value in a column and when there is none just skip it or apply np.nan. However .idmax() causes issues on some columns where it is int64 or a string. Any easy way to generate most common value for a column? thanks

def df_missing_info(df):
    most_common_value = df.apply(lambda x: x.value_counts().idxmax())

Answer 1

Because Series.value_counts also sorting Series get first index for maximal counts and add next with iter for return first value if exist else get default value, here NaN:

print (df)
   id  pts  x   y
0   1  NaN  a NaN
1   1  NaN  a NaN
2   1  NaN  s NaN
3   2  0.1  a NaN

def df_missing_info(df):
    return df.apply(lambda x: next(iter(x.value_counts().index), np.nan))
    
a = df_missing_info(df)
print (a)
id       1
pts    0.1
x        a
y      NaN
dtype: object