Summarizing Automobile Evaluation Data

Projects
1

Hi all!

Here my code of the Automobile Evaluation Data project, feel free to correct me in the comments!

import pandas as pd
import numpy as np

car_eval = pd.read_csv(‘car_eval_dataset.csv’)
print(car_eval.head())

manufacturer_country = car_eval.manufacturer_country.value_counts(normalize = True)
print(manufacturer_country)
manufacturer_no_value = car_eval[“manufacturer_country”].value_counts()/len(car_eval[“manufacturer_country”])
print(manufacturer_no_value)

buying_cost = car_eval[“buying_cost”].unique()
print(buying_cost)

buying_cost_categories = [‘low’, ‘med’, ‘high’, ‘vhigh’]

car_eval[“buying_cost”] = pd.Categorical(car_eval[“buying_cost”], buying_cost_categories, ordered=True)

median = np.median(car_eval[“buying_cost”].cat.codes)
print(median)
median_category = buying_cost_categories[int(median)]
print(median_category)

luggage = car_eval.luggage.value_counts(normalize = True)
print(luggage)
luggage_nodrop = car_eval.luggage.value_counts(dropna = False, normalize = True)
print(luggage_nodrop)

luggage_no_value = car_eval[“luggage”].value_counts()/len(car_eval[“luggage”])
print(luggage_no_value)

doors = (car_eval[“doors”] == ‘5more’).sum()
print(doors)

doors_mean = (car_eval[“doors”] == ‘5more’).mean()
print(doors_mean)

Cheers!
Jorge.

2 Likes
2

Hi Jorge,

I just found this skill path and am doing this evaluation today. Your code looks good. I could not get numpy to run on the Codecademy platform, so that is a bummer. Could you get it to work or is that why you are posting it here?

1 Like
3

Hey @nfry2672241086 !

I totally made it work, didn’t have any trouble with numpy.
I just wanted to share my code because I wasn’t able to find the project solved in the forum, thus to help others.

See ya.

4

Don’t know if this is relevant anymore BUT I did notice in my codecademy platform they did not have ‘import numpy as np’ at the top of this exercise so if you just add that it should work fine!

5

Dear all,

does anyone know if the data set for the car project is available as a csv file somewhere? Codeacademy extended the original data set by one variable for country and I would like to work on jupyter notebook for this project using the csv file:)

6

And here is my code

import pandas as pd
import numpy as np

car_eval = pd.read_csv('car_eval_dataset.csv')
print(car_eval.head())

# Table of Frequencies
freq = car_eval.manufacturer_country.value_counts()
print(freq)

# Modal: Most occuring country
modal_cat =freq.index[0]
print(modal_cat) # Output: Japan

print(freq.index[5]) # France

# 2  Table of Proportions
prop = car_eval.manufacturer_country.value_counts(normalize=True)
print(prop) # 23% for Japan

# 3 List of possible values
print(car_eval['buying_cost'].unique()) # ['vhigh' 'med' 'low' 'high']
 
# 4 Ordered list for ordinal categorical variable
buying_cost_categories = ['low', 'med' ,'vhigh' , 'high']

# 5 Convert to Type category 
car_eval['buying_cost'] = pd.Categorical(car_eval['buying_cost'], buying_cost_categories, ordered = True)
print(car_eval['buying_cost'].head())

# 6 Median for buying cost category
median_category_num = np.median(car_eval['buying_cost'].cat.codes)
print(median_category_num) 
 
median_category = buying_cost_categories[int(median_category_num)]
print(median_category) # 1

# Summarize luggage capacity
# 7 Table of proportions
prop2 = car_eval.luggage.value_counts(normalize=True)
print(prop2) 

# 8 Account for missing data
prop3 = car_eval.luggage.value_counts(dropna = False, normalize=True)
print(prop3) # no missings

print(car_eval.luggage.isnull().any()) # no missings

# 9 
prop4 = car_eval.luggage.value_counts(dropna = False)
print(prop4) 

# Passanger Capacity
count = (car_eval.doors == '5more').sum()
print(count)

# 11 
proportion = (car_eval["doors"] == '5more').mean()
print(proportion)

Happy Coding

7

sorry, but do we have now another task or why did you use it print(freq.index[5]) # France for the 4th the most frequently?