Assigning an average value

This is also one of the common approaches because of its simplicity. In the case of a numerical feature, you can just replace the missing values with the mean or median. You can also use this approach in the case of categorical variables by assigning the mode (the value that has the highest occurrence) to the missing values.

The following code assigns the median of the non-missing values of the Fare feature to the missing values:

# handling the missing values by replacing it with the median fare
df_titanic_data['Fare'][np.isnan(df_titanic_data['Fare'])] = df_titanic_data['Fare'].median()

Or, you can use the following code to find the value that has the highest occurrence in the Embarked feature and assign it to the missing values:

# replacing the missing values with the most common value in the variable
df_titanic_data.Embarked[df_titanic_data.Embarked.isnull()] = df_titanic_data.Embarked.dropna().mode().values