Issue calling to_string with float_format on Pandas DataFrame

Question

When using pandas DataFrame, I can do to_string(float_format='%.1f') on a DataFrame. However, when applying the same method to df.describe(), it failed.

The issue is self-explanatory with the following code.

>>> df = pd.DataFrame([[1, 2, 'March'],[5, 6, 'Dec'],[3, 4, 'April'], [0, 1, 'March']], columns=['a','b','m']) 
>>> df
   a  b      m
0  1  2  March
1  5  6    Dec
2  3  4  April
3  0  1  March
>>> df.to_string(float_format='%.1f')
u'   a  b      m\n0  1  2  March\n1  5  6    Dec\n2  3  4  April\n3  0  1  March'
>>> df.describe().to_string(float_format='%.1f')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Library/Python/2.7/site-packages/pandas/core/frame.py", line 1343, in to_string
    formatter.to_string()
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 511, in to_string
    strcols = self._to_str_columns()
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 439, in _to_str_columns
    fmt_values = self._format_col(i)
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 693, in _format_col
    space=self.col_space
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 1930, in format_array
    return fmt_obj.get_result()
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 1946, in get_result
    fmt_values = self._format_strings()
  File "/Library/Python/2.7/site-packages/pandas/core/format.py", line 2022, in _format_strings
    fmt_values = [self.formatter(x) for x in self.values]
TypeError: 'str' object is not callable

Answer 1

It's working in your first time because none of your types are float. You could check that with df.dtypes:

In [37]: df.dtypes
Out[37]: 
a     int64
b     int64
m    object
dtype: object

From docs:

float_format : one-parameter function, optional
formatter function to apply to columns’ elements if they are floats, default None. The result of this function must be a unicode string.

So you need to pass a function not a string:

df.describe().to_string(float_format=lambda x: '%.1f' % x)

or with .format:

df.describe().to_string(float_format=lambda x: "{:.1f}".format(x))