Predict which #TidyTuesday Scooby Doo monsters are REAL with a tuned decision tree model | Julia Silge

[utterances-bot]

Predict which #TidyTuesday Scooby Doo monsters are REAL with a tuned decision tree model | Julia Silge

A data science blog

https://juliasilge.com/blog/scooby-doo/

[RaymondBalise]

Hi Julia, This is a nearly perfect example for teaching. All it is missing is the rendering of the final decision tree. I am thinking that I will recommend students watch this and then I can share something like final_fit |> extract_fit_engine() |> rpart.plot::rpart.plot() and then show the crosswalk between the parttree plot and the traditional tree. Is that code snippet/pipeline how you would extract/show the tree? Have you recorded a demo where you show a final tree? I am super curious to know how you show (style the aesthetics) of decision trees.

[juliasilge]

@RaymondBalise Yes, that is how I would go about it! Here are two options for visualizing the decision tree:

• my blog post, which uses parttree
• Emil's ISLR labs, which uses rpart.plot