How to convert the values of an attribute having categorical values to integer type?

Question

I have a dataset in which one of its columns is Ex-Showroom_Price, and I'm trying to convert its values to integers but I'm getting an error.

import pandas as pd

#reading the dataset
cars = pd.read_csv('cars_engage_2022.csv')

cars["Ex-Showroom_Price"]  = int(cars["Ex-Showroom_Price"] .split()[-1].replace(',',''))

Error:

TypeError                                 Traceback (most recent call last)
<ipython-input-40-d65bfedf76a4> in <module>
----> 1 cars["Ex-Showroom_Price"]  = int(cars["Ex-Showroom_Price"] .split()[-1].replace(',',''))

TypeError: 'int' object is not subscriptable

Values of Ex-Showroom_Price:

Rs. 2,92,667
Rs. 2,36,447
Rs. 2,96,661
Rs. 3,34,768
Rs. 2,72,223
:

Answer 1

First split string into list.

df["cars-list"] = df['Ex-Showroom_Price'].str.split()

Then remove commas (',').

df["cars-int"] = df["cars-list"].apply(lambda x: x[-1].replace(',','') )

Then convert into int.

df["cars-int"] = df["cars-int"].astype(int)

Answer 2

You are trying to use str methods over an array of data. Assuming your cars is a DataFrame, you could try with methods iterating over single cells. str comes in handy for DataFrames

data = ["Rs. 2,92,667", "Rs. 2,36,447", "Rs. 2,96,661", "Rs. 3,34,768", "Rs. 2,72,223"]
cars = pd.DataFrame(data, columns=['Ex-Showroom_Price'])

cars["Ex-Showroom_Price"] = cars["Ex-Showroom_Price"].str.replace(r'.* ([\d,]+)+$', r'\1').str.replace(',', '').astype('int32')

I've used a regular expression here and I've kept your ',' substitution for simplicity, but you may merge them in one.

Note: the above code runs well, but as @martineau pointed out, the error you are getting seems to be related to the format of your data. Please, ensure data conforms with the format I'm assuming here, or expand you question with further details.