According to the documentation (https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.to_csv.html), we can specify the floating number format in to_csv method of dataframe.
I try to change the float format to 2 decimal percentage with float_format=':,.2%' parameter, which works on python console (like '{:,.2%}'.format(3.14)), but to_csv gives "incomplete format" error.
Use the old style formatting:
df = pd.DataFrame(np.random.random((5,5)))
df.to_csv('test.out', float_format='%.2f', index=False, header=False)
!type test.out
Output:
0.10,0.90,0.65,0.78,0.70
0.03,0.45,0.75,0.92,0.94
0.49,0.64,0.47,0.28,0.50
0.48,0.09,0.86,0.33,0.55
0.37,0.85,0.97,0.19,0.68
You could multiply your data by 100:
df = pd.DataFrame(np.random.random((5,5)))*100
df.to_csv('test.out', float_format='%.0f%%', index=False, header=False)
!type test.out
Output:
48%,73%,30%,4%,54%
76%,53%,58%,41%,22%
97%,44%,58%,59%,60%
95%,85%,47%,67%,88%
4%,73%,66%,70%,97%