Soon notebooks of different techniques used to transfer machine learning model for performance prediction of Linux Kernel size, from one kernel version to another.
Boilerplate notebook is used to load the datasets and perform feature alignment.
Feature ranking list notebook is used to create the feature_ranking_list.json files, us for feature selection.
Dataset inspection notebook measures how representative are each value for each feature in the datasets.
Model reuse notebook measures how much a model trained on 4.13 dataset can perform on 4.15 dataset.