Full Tidymodels Workflowsets Machine Learning Tutorial | Interview w/ Julia Silge (Rstudio)

📖 Learning Labs PRO (get code & #shiny app): https://university.business-science.io/p/learning-labs-pro

😀 ABOUT: This is a 2-PART LEARNING LAB.

In part 1, we *interview* @JuliaSilge , an expert in tidymodels, author of tidytext, and data scientist at RStudio. Julia has had an eclectic career path to become a data scientist at RStudio. Julia shares her data science journey.

In part 2, we do a *full tutorial* on how to use the new #Tidymodels Workflowsets #R package. #Tidymodels is a rapidly growing ecosystem for Machine Learning in R. In this FULL TUTORIAL we share how to use the new #workflowsets package for hyperparameter tuning 9 combinations of models and preprocessing recipes.

=======

Table of Contents

✨PART 1: INTERVIEW WITH JULIA SILGE, DATA SCIENTIST @RSTUDIO

00:00 Tidymodels Workflowsets | Special Guest: Julia Silge
04:26 Part 1 - Interview with Julia Silge, Data Scientist at RStudio
04:55 - Julia's Career Path: Starting in Astronomy
05:40 - First coding language was in C (R came much later)
09:10 - Path from Astronomy to RStudio (Data Science)
11:00 - Astronomists were getting jobs as data scientists
12:46 - Upskilling with MOOCs to Gain Employment
14:46 - Learning R & Tidyverse (Loved R)
16:20 - Making Exciting Projects to Gain Employment
18:10 - Problem: Text Analysis had Limits | Solution: Tidytext!
20:22 - Venturing into Open Source - Tidytext was born!
21:20 - Tidytext now has 1.4M Downloads!!!
21:50 - Getting a job with RStudio
23:00 - Learning from Real-World | Taking this experience to Rstudio
25:00 - Working on the Tidymodels Team with Max Kuhn & Company
26:00 - What is Tidymodels? https://www.tidymodels.org/
28:00 - Free Resouce #1: Tidy Textmining Book! https://www.tidytextmining.com/
29:10 - Free Resource #2: Tidy Modeling with R https://www.tmwr.org/
30:45 - Free Resource #3: Supervised Modeling for Text Analysis in R https://smltar.com/
34:19 - Where to learn more about Julia: https://twitter.com/juliasilge, https://juliasilge.com/blog/, http://youtube.com/juliasilge

✨PART 2 - FULL TIDYMODELS MACHINE LEARNING TUTORIAL W/ WORKFLOWSETS

35:13 - Part 2: Business Problem: Executive Compensation Analysis
38:25 - Shiny App Demo: CEO Compensation Explorer
41:47 - Why Workflowsets? Model Tuning for Many Models.
45:56 - Code Demo: Workflowsets | CEO Compensation Model
46:00 - Project Setup
49:21 - CEO Analysis
51:57 - Collect Data (CEO Data & Stock Data)
1:02:28 - Modeling
1:02:28 - Recipes x3 (No missing, Mean Impute, KNN Impute)
1:05:30 - Model Specs x3 (GLMNET, XGBOOST, SVM)
1:06:06 - 5-Fold Cross Validation
1:06:20 Workflowsets
1:12:52 Select & Finalize Best Model
1:14:15 Variable Importance
1:15:54 Learning More
1:23:48 Q&A