Pandas new column based on row values

Question

I have a dataframe:

    Item    SW_test HW_test QA_test
0   PC      Pass    Pass    Pass
1   Laptop  Fail    Fail    Pass
2   Mouse   Pass    Pass    Fail

I want to create a final column which will give Pass if all tests were pass (not case sensitive) and Fail if one or more are fail.

    Item    SW_test HW_test QA_test Final
0   PC      Pass    Pass    Pass    Pass
1   Laptop  Fail    Fail    Pass    Fail
2   Mouse   Pass    Pass    Fail    Fail

How can I create df['Final'] in pandas?

Space Impact · Accepted Answer · 2020-08-04T00:43:32.693

6

Use eq with all:

df['Final'] = df.iloc[:,1:].eq('Pass').all(1)
#If case sensitive you can use
df['Final'] = df.iloc[:,1:].isin(['Pass','pass']).all(1)
#or
df['Final'] = df.iloc[:,1:].apply(lambda x: x.str.lower().eq('pass')).all(1)
#or
df['Final'] = df.iloc[:,1:].applymap(str.lower).eq('pass').all(1)

Also instead of mapping again True/False using map, you can use np.where:

df['Final'] = np.where(df['Final'], 'Pass', 'Fail')

edited Aug 04 '20 at 00:43

answered Aug 03 '20 at 15:41

Space Impact

13,085
23
48

almost close to what i need. two things missing: - the result should be Pass/Fail. i can accomplish that using map - the formula only looks for Pass and when it's pass then it gives False. In my question i asked for non-case sensitive. is there a way that can be achieved? – Adnan Hadi Aug 04 '20 at 00:08
@AdnanHadi Please check the update. Also Accept and upvote if it solves your problem. – Space Impact Aug 04 '20 at 00:37

score 4 · Answer 2 · answered Aug 03 '20 at 15:43

4

cols = ['SW_test', 'HW_test', 'QA_test']
df['Final'] = df[cols].eq('Pass').all(1)

    Item    SW_test HW_test QA_test Final
0   PC      Pass    Pass    Pass    Pass
1   Laptop  Fail    Fail    Pass    Fail
2   Mouse   Pass    Pass    Fail    Fail

answered Aug 03 '20 at 15:43

Kenan

13,156
8
43
50

score 1 · Answer 3 · answered Aug 03 '20 at 15:35

1

You can apply lambda function to check where the condition holds and then can replace the true/false value with whatever you want. For example:

#create a dataframe
df = pd.DataFrame({'a':['Pass','Pass'], 'b':['Pass','Fail']})


    a       b
0   Pass    Pass
1   Pass    Fail

Create a new column where the condition holds

df['c'] = df.apply(lambda row: row.a=='Pass' and  row.b=='Pass', axis=1)


    a       b       c
0   Pass    Pass    True
1   Pass    Fail    False

Replace the true/false values with what you want to display

df['c'] = df['c'].map({ True: 'Pass', False: 'Fail'})


    a       b       c
0   Pass    Pass    Pass
1   Pass    Fail    Fail

answered Aug 03 '20 at 15:35

Maleehak

671
8
17

1

I would advise against using apply for such row like operations. The process will be very slow. Read more [here](https://stackoverflow.com/questions/54432583/when-should-i-ever-want-to-use-pandas-apply-in-my-code) – Kenan Aug 03 '20 at 15:45
Thank you for sharing this. Learnt something new today :) – Maleehak Aug 03 '20 at 15:49

score 0 · Answer 4 · answered Aug 03 '20 at 15:40

0

# set column to "Pass" initially
df["Final"] = "Pass"

# set "Fail" rows
df.loc[(
    (df.loc[:, ["SW_test", "HW_test", "QA_test"]] == "Fail") | 
    (df.loc[:, ["SW_test", "HW_test", "QA_test"]] == "fail")
  ).any(axis = 1), "Final"] = "Fail"

answered Aug 03 '20 at 15:40

Bruno Carballo

1,156
8
15

Pandas new column based on row values

4 Answers4