Previously, we have trained a really simple model that was predicting income level between “<= 50K” and “> 50K”. It was a very simple classification problem for which we could use Sklearn-porter to generate Ruby code based on the trained model.

This time I decided to pick a more complicated dataset for regression problem. To complicate the project a little bit more and make it look more like “real-life”, we implement Sklearn’s pipeline that will transform the data into the shape expected by an estimator.

Let’s implement PalmSpringsBnB that is able to suggest a list of prices based on information about the property.