Interpretable Black Box Models in R

In this week's #TidyTuesday video, I go over how to use Broom and create coefficient and effect plots for analyzing linear models. I then use #TidyModels to create a RandomForest and XGBoost model. Using the iml package, I show how to create a Variable Importance plot and how to interpret the results. I then show how to create Accumulated Local Effect plots (ALE) to understand how the Black Box models learned and predict data. I then compare the two plots to understand if the models could be stacked and how they viewed the data differently. Finally, I create a local surrogate model using LIME to explain model predictions and its importance in explaining individual predictions. #DataScience Connect with me on LinkedIn: https://www.linkedin.com/in/andrew-couch/ Q&A Submission Form: https://forms.gle/6EzU4GCR9VnJx8gg7 Code for this video: https://github.com/andrew-couch/Tidy-Tuesday/blob/master/Season%201/Scripts/TidyTuesdayInterpretiveBlackBoxModels.Rmd Interpretable Machine Learning: https://christophm.github.io/interpretable-ml-book/ TidyTuesday: https://github.com/rfordatascience/tidytuesday