How to change CSV in Python and Pandas?

Question

I am a beginner in Python and need a help.

I have a csv file that has only one column with age data.

I need to transform the integers into ranges, like "ate_21_anos", "ate_24_anos", etc.

The problem is that I cannot compare an int and return a string.

If possible, I would like to obtain this result using pandas.

So far, I’ve tried this way:

import pandas as pd
dados = pd.read_csv('Alunos.csv', delimiter=';', usecols=['IDADE_INGRESSO']
for x in dados:
   if x <= 21:
   return "menor_21"
dados

I know you’re incomplete and wrong, because I’m really new.

Below is a sample of the base I’m using:

Here are the results I want to get:

Answer 1

One way to do with pandas is to use the function apply().

df['Intervalo']=df['IDADE_INGRESSO'].apply(lambda x: 'menor_21' if x<21 else ('menor_24' if x<24 else 'maior_24'))

If your rule is more complicated, such as wanting an interval of 18, 21, 24,... it is also possible to create a decision function and apply with the apply().

def define_intervalo(num):
    for faixa in [21,24]:
        if num < faixa:
            return 'menor_{}'.format(faixa)
    return 'maior_{}'.format(faixa)

df['Intervalo'] = df['IDADE_INGRESSO'].apply(define_intervalo)

Answer 2

An idea..

You can use the pandas cut command.

To demonstrate how you use, instead of importing a list, I did it with a list.

import pandas as pd

idades = [21, 33, 15, 21, 28, 60, 35, 19, 41, 10, 18, 38, 22,]
bins = [0, 21, 24, 100]
idades_ingresso = pd.cut(idades, bins)
idades_ingresso

[(0, 21], (24, 100], (0, 21], (0, 21], (24, 100], ..., (24, 100], (0, 21], (0, 21], (24, 100], (21, 24]] Length: 13 Categories (3, interval[int64]): [(0, 21] < (21, 24] < (24, 100]]

04 categories/ tracks were created

idades_ingresso.categories

Intervalindex([(0, 21], (21, 24], (24, 100]], closed='right',dtype='interval[int64]')

You could count the values that are in track record, which in your case, would be the ages.

pd.value_counts(idades_ingresso)

(24, 100] 6

(0, 21] 6

(21, 24] 1

dtype: int64

Answer 3

You can use the function np.where() for that reason.

Supposing you have a spine idade within a pandas Dataframe by name df:

import numpy as np

df['faixa_etaria'] = np.where(df.idade<=21,'até_21',np.where(df.idade<=24,'entre 21 e 24',np.where(df.idade<=35,'entre 24 e 35','mais de 35')))

This way it will have 3 age groups with this conditional chained!