Для одного столбца (серии) строк вам не нужно astype(str)
:
In [518]: df = pd.Series(['one','two','three'])
In [519]: df
Out[519]:
0 one
1 two
2 three
dtype: object
In [520]: df.values
Out[520]: array(['one', 'two', 'three'], dtype=object)
In [521]: [len(i) for i in df.values]
Out[521]: [3, 3, 5]
In [522]: max([len(i) for i in df.values])
Out[522]: 5
In [523]: df.values.astype(str)
Out[523]: array(['one', 'two', 'three'], dtype='<U5')
In [524]: [len(i) for i in df.values.astype(str)]
Out[524]: [3, 3, 5]