pandas mapping function to transform data using dictionary - pandas

Question

please help friends.
I want to use mapping to match the age of student and identify them in category adult or child by comparing it with a dictionary 'dlist' containing age 1 to 18 as child and age 19 to 60 as adult..
# making Data Frame
age=np.random.randint(1,50,5,int)
name=['kashif', 'dawood', 'ali', 'zain', 'hamza']
df5=pd.DataFrame({'name':name,
'age':age})
# making dictionary
dlist={range(1,18):'child' , range(19,50):'adult'}
# now maping dictionary with data frame 'age' column elements to add status adult if age greater than 18 using dictionary
df5['Status']=df5.age.map(dlist)
but it returns the data frame with column name 'Status' but NAN values (instead of adult or child)
kindly ignore my English if there are mistakes. i m not a native speaker of english.

Answer 1

In Python 3, you are allowed to use ranges as dict-keys, but it does not work the way you seem to think. For example
print(dlist[1])
will give you a key-error, since the key 1 does not exist in dlist, however
print(dlist[range(1,18)])
will work, since you have a key that is range(1,18). This means that you can not use your dlist the way you want in the map-function
To make use of your dict, with ranges as keys, you should instead use apply
df5['Status'] = df5['age'].apply(
lambda x: next((v for k, v in dlist.items() if x in k), 'NA')
)
Where [v for k, v in dlist.items() if x in k] gives you a list of all values in your dict, if x is in k (which is a range). The next()-function gets the next value (i.e. the first value) in that list (but it also works on iterators, and thats why [] can be omitted. NA is the default value for next() if no next exists. See https://docs.python.org/3/library/functions.html#next
You should howevr note that range(1,18) does NOT include 18. So with this code, the age 18 will give you Status = 'NA'

Answer 2

you can achieve by using np.where
df5['status'] = np.where((df5['age']>=1) & (df5['age']<=18), 'child', 'adult')
print(df5)
name age status
kashif 15 child
dawood 11 child
ali 33 adult
zain 21 adult
hamza 31 adult

Answer 3

That's my personal preference when working with pandas. I always use cut() method of pandas with a list of labels and bins to create a categorical variable:
import numpy as np
import pandas as pd
# making Data Frame
np.random.seed(41)
age=np.random.randint(1,50,5,int)
name=['kashif', 'dawood', 'ali', 'zain', 'hamza']
df=pd.DataFrame({'name':name, 'age':age})
# create a bin
bins = [0, 18, 50]
# create a bin label
label_list = ['adult', 'old']
# create a new column with bin and label
df['status'] = pd.cut(df.age, bins, labels=label_list)

Answer 4

use np.select
#specify conditions
conditions=[(df5['age']<=18),
(df5['age']>18)& (df5['age']<=50)]
#specify column output based on conditions
choices = ['child','adult'] #you can also specify numbers as well here
#create status column based on conditions
df5["status"] = np.select(conditions, choices)