Supporting data and code for the manuscript "Virtual borehole formation prediction using stacked machine learning algorithms".

We use drilling data from a certain area along the coast of China for stratigraphic coding and normalized data pro- cessing, and then use the previously mentioned method to train predictions and obtain better results. Section 4 discusses the similarity and error of 10-fold cross-validation of real borehole and predicted borehole data. Section 5 summarizes the full text. This study proposes a new prediction method based on machine learning to address existing prob- lems in the field of virtual borehole formation prediction, such as insufficient prediction accuracy, low automation, and low efficiency.