While we would love to obtain datasets that contain no missing values whatsoever, the reality is that we almost always have to handle them. This recipe shows you four methods that you can use.
Pandas provides a fillna() method to fill in missing values. Create a DataFrame from the customer data using the previous recipe, and then try each of the following methods.
The following code tells us how to fill in missing values in Pandas:
# 1: Replace all missing values with a string - 'Missing'
customers.fillna('Missing', inplace=True)
# 2: Replace all missing values with a 0
customers.fillna(0, inplace=True)
# 3: Replace all missing values with the mean of the DataFrame
customers.fillna(raw_data.mean(), inplace=True)
# 4. Replace the missing values of a single colum with the mean of that column
customers['marketing_score'].fillna(raw_data.mean()['marketing_score'], inplace=True)