how to add column in dataframe based on some condition from another column?

Question

I have a python data frame.

ID              status

5                  
from 4.3 to 5   yes   
from 6 to 7.2   yes
6.3
6

I want to add another column col, as if status is missing then corresponding value is ID else substring(first integer in string) of ID

result should be like this

ID              status    col

5                         5
from 4.3 to 5   yes       4.3
from 6 to 7.2   yes       6
6.3                       6.3
6                         6

sorry for very bad representation of the question.

BENY · Accepted Answer · 2020-01-26 17:03:29Z

3

Using findall

df.ID.str.findall('[-+]?\d*\.\d+|\d+').str[0]
0      5
1    4.3
2      6
3    6.3
4      6
Name: ID, dtype: object

answered Jan 26, 2020 at 17:03

BENY

324k22 gold badges176 silver badges250 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Andrej Kesely · Accepted Answer · 2020-01-27 10:13:52Z

2

Another method, if you don't prefer regular expressions:

df['col'] = df['ID'].apply(lambda x: x if len(str(x).split()) == 1 else str(x).split()[1])
print(df)

              ID status  col
0              5           5
1  from 4.3 to 5    yes  4.3
2  from 6 to 7.2    yes    6
3            6.3         6.3
4              6           6

edited Jan 27, 2020 at 10:13

answered Jan 26, 2020 at 17:07

Andrej Kesely

196k15 gold badges60 silver badges105 bronze badges

2 Comments

Mohd Bilal Over a year ago

'float' object has no attribute 'split'. How to slove this problem of type??

Andrej Kesely Over a year ago

@MohdBilal To convert to string use str() function. I updated my answer

Bharath_Raja · Accepted Answer · 2020-01-26 17:01:01Z

1

you can loop and check over

import re

df['col'] = None

for i in range(len(df)):
    if df.loc[i, 'status'] == np.NaN:
         df.loc[i, 'col'] = df.loc[i, 'ID']
    else:
         df.loc[i, 'col'] = re.findall(r'\d+', df.loc[i, 'ID'])[0]

answered Jan 26, 2020 at 17:01

Bharath_Raja

6608 silver badges18 bronze badges

Comments

ksh22 · Accepted Answer · 2020-01-26 17:58:51Z

1

def fun(x,y):
  return (x.split("from ")[1].split(" to ")[0] if pd.notnull(y) else x)

df["sep"]=df.apply(lambda x: fun(x["ID"],x["status"]),axis=1)
df

    ID             status   sep
0   5              None     5
1   from 4.3 to 5   yes     4.3
2   from 6 to 7.2   yes     6
3   6.3            None     6.3
4   6              None     6

Assuming ID column is string

answered Jan 26, 2020 at 17:58

ksh22

437 bronze badges

Comments

zappee · Accepted Answer · 2020-01-26 22:04:22Z

0

this is correct, please check it:

else:
     df.loc[i, 'col'] = re.findall(r'\d+', df.loc[i, 'ID'])[0]

edited Jan 26, 2020 at 22:04

zappee

23.1k15 gold badges97 silver badges155 bronze badges

answered Jan 26, 2020 at 17:10

mohit kaushik

263 bronze badges

Collectives™ on Stack Overflow

how to add column in dataframe based on some condition from another column?

5 Answers 5

Comments

2 Comments

Comments

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

5 Answers 5

Comments

2 Comments

Comments

Comments

Comments

Related