how to remove scientific notation in pandas from column object type

Question

I have df with below data which I am saving in csv file :

    A      B
1   ABC    0.00772998456623635
2   XYZ    -6.745157813050465e-05
3   PQR    UDS
5   NA  
6   TES 
7   SEZ    0.051751390215281516

Column B is of object type and some number are showing as scientific notation format. I wanted to suppress scientific notation. I tried below methods but nothing worked :(

np.set_printoptions(suppress=True)

df.to_csv('test.csv',sep='|',index=False, header=False,float_format='%f')

pd.set_option('display.float_format', '{:.60g}'.format)

Can any one help me How I can change number format from -6.745157813050465e-05 to -0.0000674515781305046

Answer 1

Interesting question. Pandas astype can't manage this, and if you use errors='ignore' it won't change it. It works with a simple custom func:

def to_float_custom(x):
    if not pd.isna(x):
        try:
            return float(x)
        except ValueError:
            pass
    return x

df.B = df.B.apply(to_float_custom)

Then you have it as float, and can use pd.set_option('display.float_format', '{:.60g}'.format) to get it as a decimal:

df

    A   B
1   ABC     0.007729984566236349982637499067550379550084471...
2   XYZ     -6.74515781305046527565466574216657136275898665...
3   PQR     UDS
5   NaN     None
6   TES     None
7   SEZ     0.051751390215281516116174742592193069867789745...