Apply conditional statement to specific rows in Python

Question

I have a dataset where I would like a certain statement to only be applied to specific rows in my dataset. I would like this to only be applied to rows that contain [type] == 'aa', and apply second statement for rows that contain [type] == 'bb'

Data

location    type    mig1        de          mig2        re
ny          aa      8/1/2021    10/1/2021   1/1/2022    2/1/2022
ny          aa      8/1/2021    10/1/2021   1/1/2022    2/1/2022
ca          aa      8/1/2021    10/1/2021   1/1/2022    2/1/2022
tx          bb      9/1/2021    11/1/2021   2/1/2022    3/1/2022

Desired

 #Date 8/1/2022 is input, which yields:

    location    type    mig1        de          mig2        re
    ny          aa      03/1/2022   05/1/2022   8/1/2022    7/1/2022
    ny          aa      03/1/2022   05/1/2022   8/1/2022    7/1/2022
    ca          aa      03/1/2022   05/1/2022   8/1/2022    7/1/2022
    tx          bb      03/1/2021   11/1/2021   08/1/2022   3/1/2022



Logic - we see that the shift applies to all columns when the type == aa
        when the type value == bb, a different statement is applied

Doing

#takes input value

datevalue = pd.to_datetime(input("Enter shift: "))

#shifts dates from the datevalue input - However I would like d variable to only be applied to the rows that contain [type] == 'aa'

#apply e variable to rows that contain [type] = 'bb'

d = {
    'mig1': pd.DateOffset(months=5),
    'de': pd.DateOffset(months=3),
    're': pd.DateOffset(months=1),
}

e = {
        'mig1': pd.DateOffset(months=5),
        
    }


s = pd.Series(d).rsub(datevalue)
df.assign(**{**s, 'mig2': datevalue})

Any suggestion is appreciated

Sounds a lot like this question.

edd
– edd

2021-12-29 00:09:36 +00:00
Commented Dec 29, 2021 at 0:09 — edd
– edd, Commented Dec 29, 2021 at 0:09

Corralien · Accepted Answer · 2021-12-29 16:25:30Z

1

Try:

datevalue = pd.to_datetime(input("Enter shift: "), dayfirst=False)  # 08/01/2022

new_values = pd.Series(d).rsub(datevalue) \
                         .append(pd.Series({'mig2': datevalue})) \
                         .dt.strftime('%-m/%-d/%Y')
df.update(pd.DataFrame([new_values], index=df[df['type'] == 'aa'].index))

# do the same for type='bb' and 'e' dict
new_values = pd.Series(e).rsub(datevalue) \
                         .append(pd.Series({'mig2': datevalue})) \
                         .dt.strftime('%-m/%-d/%Y')
df.update(pd.DataFrame([new_values], index=df[df['type'] == 'bb'].index))

Output:

>>> df
  location type      mig1         de      mig2        re
0       ny   aa  3/1/2022   5/1/2022  8/1/2022  7/1/2022
1       ny   aa  3/1/2022   5/1/2022  8/1/2022  7/1/2022
2       ca   aa  3/1/2022   5/1/2022  8/1/2022  7/1/2022
3       tx   bb  3/1/2022  11/1/2021  8/1/2022  3/1/2022

edited Dec 29, 2021 at 16:25

answered Dec 29, 2021 at 11:06

Corralien

121k8 gold badges43 silver badges68 bronze badges

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

15 Comments

Lynn Over a year ago

Ok will try- how did you get the final output? Did you concatenate the ‘d’ and ‘e’ output?

Corralien Over a year ago

I understood now mig2 is datevalue.

Corralien Over a year ago

Did you have time to try my answer? I also answered to your (similar) question

Lynn Over a year ago

Will try- I marked other soln correct as I see how the output will be derived thank you. I like your explanation as well

Corralien Over a year ago

It's to avoid horizontal scrollbar. You can remove them to reformat the code on one line.

|

wwnde · Accepted Answer · 2021-12-29 00:10:12Z

1

#Coerce dates to datetime
df1=df1.set_index(['location','type']).apply(lambda x: pd.to_datetime(x,format='%d/%m/%Y'))

#Set non dates as index, slice level two and impose the datetifference
df1.loc[ ( slice(None), 'aa' ),  : ]=df1.loc[ ( slice(None), 'aa' ),  : ]-pd.to_timedelta(5, unit='d')


                  mig1         de       mig2         re
location type                                            
ny       aa   2021-01-03 2021-01-05 2021-12-27 2021-12-28
         aa   2021-01-03 2021-01-05 2021-12-27 2021-12-28
ca       aa   2021-01-03 2021-01-05 2021-12-27 2021-12-28
tx       bb   2021-01-09 2021-01-11 2022-01-02 2022-01-03

answered Dec 29, 2021 at 0:10

wwnde

26.7k6 gold badges22 silver badges38 bronze badges

7 Comments

Lynn Over a year ago

hi- was wondering if input statement is used? - needing to input a specific date

wwnde Over a year ago

Not sure I understood your question. Input what date and how?

Lynn Over a year ago

Ok. I am using input() function to input the date : 8/01/2022- whenever a specific date is entered the dates will shift. Only wish to apply to certain rows though

wwnde Over a year ago

possible but quite a complicated model especially with the multiple date fields. Will try figure it out

wwnde Over a year ago

As I figure that out, will leave my answer just in case someone needs to use slice :-)

|

Collectives™ on Stack Overflow

Apply conditional statement to specific rows in Python

2 Answers 2

15 Comments

7 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

15 Comments

7 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related