How I used Polynomial Features to make accurate predictions on sklearn’s make_friedman2 dataset

As I near the end of sklearn’s dataset range, I have come to the make_friedman2 dataset. I had previously posted about the make_friedman1 dataset, and that post can be found here:- https://medium.com/geekculture/sklearns-make-friedman1-dataset-is-a-difficult-dataset-to-predict-on-5dfb7d970426

Sklearn’s make_friedman2 dataset is a regression problem that is described in his work, “Multivariate adaptive regression splines”, The Annals of Statistics 19 (1). The construction of the dataset is described in the screenshot below. It has four features and one label:-

I wrote the script in Google Colab, which is a free online Jupyter Notebook. The main advantage of Google Colab, as with other Google products, is that it is portable. Because this platform is online, it can be accessed from any computer that has internet access and a search engine. The only true disadvantage of Google Colab is the fact that it does not have an adequate undo function, so care needs to be taken to ensure that valuable code is not overwritten or even deleted forever.