Pandas: Insert New column to dataframe and populate values within new column based on if then logic

Question

I have a dataframe with two country descriptions. Sometimes they match, sometimes they don't.

Country Desc1        Country Desc2

1    US           US  
2    US           UK           
3    UK           US    
4    UK           UK

I need to 1.) insert another column (Country Desc3) with all the row values populated with 2.) a rule that returns Country Desc1 if it matches Country Desc2.

Possible duplicate of how to compare two columns in pandas to make a third column ? — Zero
– Zero, Commented Jul 25, 2017 at 17:02
If you need NaN I will recommend Both @piRSquared 's and @Rayhane Mama's answer — BENY
– BENY, Commented Jul 25, 2017 at 17:35

piRSquared · Accepted Answer · 2017-07-25 17:33:15Z

2

df['Country Desc3'] = \
    df['Country Desc1'].mask(df['Country Desc1'] != df['Country Desc2'])

df

  Country Desc1 Country Desc2 Country Desc3
0            US            US            US
1            US            UK           NaN
2            UK            US           NaN
3            UK            UK            UK

answered Jul 25, 2017 at 17:33

piRSquared

296k68 gold badges509 silver badges654 bronze badges

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

Comments

Scott Boston · Accepted Answer · 2017-07-25 17:04:45Z

1

Let's use iloc and join:

df['Country Desc3'] = df.apply(lambda x: x.iloc[0] if x.iloc[0] == x.iloc[1] else ', '.join(x),axis=1)

Output:

  Country Desc1 Country Desc2 Country Desc3
1            US            US            US
2            US            UK        US, UK
3            UK            US        UK, US
4            UK            UK            UK

answered Jul 25, 2017 at 17:04

Scott Boston

154k15 gold badges160 silver badges207 bronze badges

Comments

BENY · Accepted Answer · 2017-07-25 17:38:46Z

1

Try this , if you need a string in new column

df['Country Desc3']=df.apply(lambda x: ','.join(x.unique().tolist()), axis=1)

If you need a list in new column

df['Country Desc3']=df.apply(lambda x: x.unique().tolist(), axis=1)

Just in case you need NaN for no matched row

df['Country Desc3']=np.nan
df.loc[df['Country Desc1']==df['Country Desc2'],'Country Desc3']=df['Country Desc1']

edited Jul 25, 2017 at 17:38

answered Jul 25, 2017 at 17:32

BENY

324k22 gold badges176 silver badges250 bronze badges

Comments

Rayhane Mama · Accepted Answer · 2017-07-25 17:39:00Z

1

You can do that using numpy.where as follows:

df['Country Desc3'] = np.where(df['Country Desc1']==df['Country Desc2'],df['Country Desc1'],np.nan)

this will give you:

  Country Desc1 Country Desc2 Country Desc3
1            US            US            US
2            US            US            US
3            UK            US           NaN
4            UK            UK            UK

if you don't like to have NaN values just change np.nan by whatever you like, for example : df['Country Desc1']+', '+df['Country Desc2'] to have the concat of the two columns when they don't match.

edited Jul 25, 2017 at 17:39

answered Jul 25, 2017 at 17:12

Rayhane Mama

2,42413 silver badges20 bronze badges

Collectives™ on Stack Overflow

Pandas: Insert New column to dataframe and populate values within new column based on if then logic

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related