I would love to see this sort of analysis for university administrators. I think most would be quite shocked when comparing their "compensation" to the faculty for instance.
Like the fact that Julia has a rigorous background in science. I think that training is too often missing amongst those who label themselves as "Data Scientists".
Thank you! This was great inspiration. Question. I am a bit unsure about the grid = 15 argument you specify..? How does this influence my RF and XGB model tunings? When I run my workflowsets I get the following errors, that I don't really know how to interpret: [1] Warning: No tuning parameters. `fit_resamples()` will be attempted and the Error: 2 of 15 resampling: normalized_knn failed with preprocessor 1/1, model 1/1 (predictions): [2] Error: Problem with `mutate()` column `.row`.ℹ `.row = orig_rows`.ℹ `.row` must be size 710 or 1, not 723. What might be the reason? I also experienced this error message only for impute_knn_RF and impute_knn_XGB [3] Some tuning parameters require finalization but there are recipe parameters that require tuning. Please use `parameters()` to finalize the parameter ranges. What does it actually mean? Any help is helpful! Thank you!
grid = 15 is to build grid of 15 parameter combinations. I teach a lot of this stuff in my courses if you'd like to learn in depth how to do data science with R. university.business-science.io/p/5-course-bundle-machine-learning-web-apps-time-series