Learn to predict class membership for the Datasaurus Dozen dataset using tidymodels in this informative 27-minute video tutorial by Julia Silge. Explore multiclass evaluation metrics to determine which of the #TidyTuesday Datasaurus Dozen are easier or harder for a random forest model to identify. Follow along as the tutorial covers key concepts including resampling, preprocessing, evaluation, ROC curves, confusion matrices, and result visualization. Gain practical insights into model performance and interpretation, with code examples available on Julia's blog for further study and implementation.
Overview
Syllabus
Introduction
Get mean and correlation
Resampling
Preprocessing
Evaluation
Results
ROC Curves
ROC Comparison
Confusion Matrix
Visualization
Conclusion
Taught by
Julia Silge
Related Courses
-
Understand data science for machine learning
-
Build Features for Machine Learning from Netflix Description Text
-
Intermediate Data Manipulation and Machine Learning
-
Use Tidymodels Workflowsets to Predict Human-Computer Interactions from Star Trek
-
Predictive Modeling in R With Tidymodels and NFL Attendance
-
Tuning Random Forest Hyperparameters with Tidymodels
