Pandas Float decimals

Question

I`m trying to write to a csv a pandas DataFrame with a column with various data types. I need all the float values to have exactly 4 decimal places.

For example. I define a new df:

df=pd.DataFrame({'col1': [0.24567899874, 'a'], 'col2' : [0.123456789, 1] })

And then writting it to a csv file using float_format:

df.to_csv('example.csv', float_format='%.4f')

I get the result: col1, col2 0.24999999999986597,0.1234 a,1.0000

It seems that float_format only applies to columns in which all their elements are numeric values.

Answer 1

for col in df.columns:
    if df[col].dtype == 'float':
        df[col].round(4)

Answer 2

You could do this:

for col in df.columns:
    df[col] = df[col].apply(lambda x: '{:.4f}'.format(x) if type(x) is int or type(x) is float else x)

df.to_csv('example.csv')