how to convert mm:ss values in dataframe using python

I am pretty new to pandas and have a tabular data which has 3 columns where (df.info() gives object type) values are both in s and m:s. I would like to convert all m:s values to s values. I'll leave some examples to clarify.

now:

VT          FT          TTPF
1:28.8      1:17.2      30.4
1:06.4      1:06.2      16.8
38.6        26.2        10.8

should be:

VT          FT          TTPF
88.8        77.2        30.4
66.4        66.2        16.8
38.6        26.2        10.8

I'm sorry if I couldn't provide all the details. Feel free to request edits

Solution

First select only strings columns by DataFrame.select_dtypes and DataFrame.apply custom function with Series.str.contains for filter values with : and then Series.str.split with casting to floats, multiple by 60 and sum together in Series.mask for rows with ::

def f(x):
    m = x.str.contains(':')
    y = x[m].str.split(':', expand=True)
    return x.mask(m, y[0].astype(float) * 60 + y[1].astype(float))

c = df.select_dtypes(object).columns
df[c] = df[c].apply(f).astype(float)
print (df)
     VT    FT  TTPF
0  88.8  77.2  30.4
1  66.4  66.2  16.8
2  38.6  26.2  10.8

Another idea is use DataFrame.applymap for elemetwise processing:

def f(x):
    if ':' in str(x):
        a, b = x.split(':')
        return float(a) * 60 + float(b)
    else:
        return float(x)

df = df.applymap(f)
print (df)
     VT    FT  TTPF
0  88.8  77.2  30.4
1  66.4  66.2  16.8
2  38.6  26.2  10.8