Suppose that we have a website where people can sell and buy used cars. When posting an ad on our website, the sellers often struggle to come up with a meaningful price. We want to help our users with automatic price recommendation. We ask the sellers to specify model, make, year, mileage, and other important characteristics of a car, and based on that information, we want to suggest the best price.
- First, we download the dataset.
- Next, we do some preliminary analysis of the data.
- After that, we set up a validation strategy to make sure our model produces correct predictions.
- Then we implement a linear regression model in Python and NumPy.
- Next, we cover feature engineering - to extract important features from the data to improve the model
- Finally, we see how to make our model stable with regularization and use it to predict car prices.