Lecture 0 - Tidyverse
Tidyverse:
- Core Tidyverse
- Wrangle (tidyr & dplyr)
- Program (purrr)
R Code Evaluation Methods:
- Standard Evaluation
- Non-Standard Evaluation
- Tidy Evaluation
Pipe Operator:
- Basic Piping
- Argument Placeholder
- Re-using Placeholder for Attributes
Lecture 1: Tidymodels
Recipe:
- feature engineering
- steps
- recipes
Parsnip:
- engines
- modelling
- workflows
Rsample:
- sampling
- validation
- cross-validation
Lecture 2: Machine Learning, Regression & Classification
Regression Algorithms:
- Linear Regression
- Ridge, LASSO, Elastic Net
- MARS
- SVM
- KNN
- Bagging
- Random Forest
- XGBoost, LightGBM, CatBoost
- Cubist
- Multi Layer Perceptron
Classification Algorithms:
- Naive Bayes
- Logistic Regression
- Ridge, LASSO, Elastic Net
- SVM
- KNN
- CART
- Bagging
- Random Forest
- XGBoost, LightGBM, CatBoost
- Cubist
- Multi Layer Perceptron
Lecture 3: Hyperparameter Tuning
Tune:
- tuning
- grid searches
- validation
Lecture 4: Ensemble Learning & Stacking
Stacks:
- ensembling
- stacking
- elastic net stacking
- simple ensembles
Lecture 5: Automatic Machine Learning
H2O:
- H2O modelling
- AutoML
- Tidymodels integration, h2oparsnip