CODEWITHSUNDEEP
pythonpandas
Bharat Sharma
Bharat Sharma

Reputation: 1219

compute columns based on multiple conditions

I was reading a blog for conditaion based new computations where new col 'category' is inserted.

data = {'name': ['Jason', 'Molly', 'Tina', 'Jake', 'Amy'], 
        'age': [42, 52, 36, 24, 73], 
        'preTestScore': [4, 24, 31, 2, 3],
        'postTestScore': [25, 94, 57, 62, 70]}
df = pd.DataFrame(data, columns = ['name', 'age', 'preTestScore', 'postTestScore'])
df['category'] = np.where(df['age']>=50, 'yes', 'no')

how it can be extend to more that multiple conditions like if age is less than 20 then kid ; if between 21 and 40 then young ; if above 40 then old

Upvotes: 2

Views: 491

Answers (2)

BENY
BENY

Reputation: 323316

You can using pd.cut (BTW , 40 is not old man :-()

pd.cut(df.age,bins=[0,20,39,np.inf],labels=['kid','young','old'])
Out[179]: 
0      old
1      old
2    young
3    young
4      old
Name: age, dtype: category
Categories (3, object): [kid < young < old]

Upvotes: 1

ALollz
ALollz

Reputation: 59579

For multiple conditions, you can just use numpy.select instead of numpy.where

import numpy as np

cond = [df['age'] < 20, df['age'].between(20, 39), df['age'] >= 40]
choice = ['kid', 'young', 'old']

df['category'] = np.select(cond, choice)
#    name  age  preTestScore  postTestScore category
#0  Jason   42             4             25      old
#1  Molly   52            24             94      old
#2   Tina   36            31             57    young
#3   Jake   24             2             62    young
#4    Amy   73             3             70      old

Upvotes: 5

Related Questions