Downloading the UCI Automobile dataset

The Automobile dataset was created by Jeffrey C. Schlimmer and published in UCI's machine learning repository. It contains information about 201 automobiles, along with their prices. The names of the features are missing. Nevertheless, I could get them from the dataset's description (http://archive.ics.uci.edu/ml/machine-learning-databases/autos/imports-85.names). So, we can start by seeing the URL and the feature names, as follows:

url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/autos/imports-85.data'

header = [
    'symboling',
    'normalized-losses',
    'make',
    # ... some list items are omitted for brevity 
    'highway-mpg',
    'price',

]

Then, we use the following code to download our data.

df = pd.read_csv(url, names=header, na_values='?')

It is mentioned in the dataset's description that missing values...