Splitting dataframe based on Floating value in python-CodePudding

I tried to takeout decimal values based on value present before decimal point .I have data frame like below,

data    flow
1.5 parallel
1.2 parallel
1.3 parallel
2   sequence
2.5 parallel
2.4 parallel
2.8 parallel
3   sequence
3.2 parallel
3.1 parallel
3.5 parallel
4   sequence
4.1 parallel
4.5 parallel
4.3 parallel
1   sequence
5   sequence
6   sequence

Expected output,

data    flow
1.5 Parallel1
1.2 Parallel1
1.3 Parallel1
2   sequence
2.5 Parallel2
2.4 Parallel2
2.8 Parallel2
3   sequence
3.2 Parallel3
3.1 Parallel3
3.5 Parallel3
4   sequence
4.1 Parallel4
4.5 Parallel4
4.3 Parallel4
1   sequence
5   sequence
6   sequence

How can i achieve this using pands,...

CodePudding user response：

If data is a string:

df.loc[df['flow'].ne('sequence'), 'flow']  = df['data'].str.extract('(\d )',
                                                                    expand=False)

if it is a float:

df.loc[df['flow'].ne('sequence'), 'flow']  = df['data'].astype(int).astype(str)

output:

    data       flow
0    1.5  parallel1
1    1.2  parallel1
2    1.3  parallel1
3    2.0   sequence
4    2.5  parallel2
5    2.4  parallel2
6    2.8  parallel2
7    3.0   sequence
8    3.2  parallel3
9    3.1  parallel3
10   3.5  parallel3
11   4.0   sequence
12   4.1  parallel4
13   4.5  parallel4
14   4.3  parallel4
15   1.0   sequence
16   5.0   sequence
17   6.0   sequence