combine two columns based on a condition - python

Question

I have the following DataFrame with two columns and I would like to create a new column based on this condition :

if the value of y is -1 take the value of x
if the value of x is -1 take the value of y

df = pd.DataFrame({'x': ['day', 'day', 'night', '-1', '-1', '-1'],
                   'y': ['-1', '-1', '-1', 'night', 'day', 'day']})
df

I have tried the following but I got a None and NaN output

def func(x):
    if (x['y'] == -1):
        return x['x']
    elif (x['x'] == -1):
        return x['y']

df = df.assign(z=df.apply(func, axis=1))
df

and

conditions = [
    (df['y'] == -1),
    (df['x'] == -1),
]

choices = [df['x'],df['y']]
df['z1'] = np.select(conditions, choices, default=np.nan)
df

the expected result should be like this

df = pd.DataFrame({'x': ['day', 'day', 'night', '-1', '-1', '-1'],
                   'y': ['-1', '-1', '-1', 'night', 'day', 'day'],
                   'z':['day','day','night','night','day','day']})
df

svfat · Accepted Answer · 2022-12-11 08:32:41Z

2

What you are doing is fine, except one moment: you are comparing with integer in func, but you have strings in your dataframe, consider to rewrite it like this (add quotes around -1):

def func(x):
    if (x['y'] == '-1'):
        return x['x']
    elif (x['x'] == '-1'):
        return x['y']

or use integers in the dataframe:

df = pd.DataFrame({'x': ['day', 'day', 'night', -1, -1, -1],
                   'y': [-1, -1, -1, 'night', 'day', 'day']})

but not both at the same time of course.

You can also consider casting types or comparing with multiple values n in [-1, '-1'] if it suits your needs better.

edited Dec 11, 2022 at 8:32

answered Dec 11, 2022 at 8:27

svfat

3,3931 gold badge18 silver badges34 bronze badges

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

Collectives™ on Stack Overflow

combine two columns based on a condition - python

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related