change value in specific rows with column conditions in pandas

Question

I have a dataframe df with two columns gender, score.

|---------------------|------------------|
|      gender         |     score        |
|---------------------|------------------|
|          male       |         34       |
|---------------------|------------------|
|          female     |         34       |
|---------------------|------------------|
|          male       |         34       |
|---------------------|------------------|
|          female     |         34       |
|---------------------|------------------|
|          male       |         34       |
|---------------------|------------------|

I want to change scores of males (gender == 'male') from row 3 to row 5 to be 0, expected output:

|---------------------|------------------|
|      gender         |     score        |
|---------------------|------------------|
|          male       |         34       |
|---------------------|------------------|
|          female     |         34       |
|---------------------|------------------|
|          male       |         0        |
|---------------------|------------------|
|          female     |         34       |
|---------------------|------------------|
|          male       |         0        |
|---------------------|------------------|

How can I combine iloc with that condition?

Your question is unclear. Please include an example of the input and the expected output. — DYZ
– DYZ, Commented Jul 8, 2018 at 20:56
You should use loc not iloc here. df.loc[df.gender == 'male', 'score'] = 0 — Anton vBR
– Anton vBR, Commented Jul 8, 2018 at 21:04
@AntonvBR I don't want to change the first row while the gender is male. — Harold
– Harold, Commented Jul 8, 2018 at 21:06
Is "3 to 5" a hardwired condition? Or do you believe that you have a case of replicated data? — DYZ
– DYZ, Commented Jul 8, 2018 at 21:21

Anton vBR · Accepted Answer · 2018-07-08 21:31:41Z

1

Alt1:

You could do it with two masks (conditions). This should be readable and make sense.

m1 = (df.gender == 'male')
m2 = (df.gender.duplicated())

df.loc[m1&m2, 'score'] = 0

Alt2:

Slice away the first truth value of the nonzero mask (requires import numpy as np). This should be faster.

m = np.nonzero(df.gender=='male')[0][1:]
df.loc[m, 'score'] = 0

Full example:

import pandas as pd
import numpy as np

df = pd.DataFrame({
    'gender': ['male','female','male','female','male'],
    'score': 34
})

m1 = (df.gender == 'male')
m2 = (df.gender.duplicated())

m = np.nonzero(df.gender=='male')[0][1:]
df.loc[m, 'score'] = 0

print(df)

Returns:

   gender  score
0    male     34
1  female     34
2    male      0
3  female     34
4    male      0

edited Jul 8, 2018 at 21:31

answered Jul 8, 2018 at 21:20

Anton vBR

19k6 gold badges47 silver badges47 bronze badges

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

Comments

Pyd · Accepted Answer · 2018-07-09 05:22:24Z

0

I think you need,

m=df.loc[2:5,:].loc[df['gender']=='male']
df.loc[m.index,'score']=0
print(df)
    gender  score
0   male    34
1   female  34
2   male    0
3   female  34
4   male    0

answered Jul 9, 2018 at 5:22

Pyd

6,16919 gold badges59 silver badges117 bronze badges

Collectives™ on Stack Overflow

change value in specific rows with column conditions in pandas

2 Answers 2

Alt1:

Alt2:

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Alt1:

Alt2:

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related