Concrete strength prediction

Description

In this project, I implement a Generalized Additive Model (GAM) to do regressions and predict the strength of concrete based in at least the 4 more correlative features with the strength in the dataset. Along the project I create 4 models to show the main difference in the metrics when scaling and no-scaling the features data. I also apply the GridSearch method from the LinearGAM to check different $\lambda$ values.

The results in this project are summarized at the end of the jupyter notebook where the non-scaled simple GAM model shows an slightly better performance than the ones with GridSearch.

Author

Dr. Raúl Guerrero