4T-Instructional Series in Machine Learning and Artificial Intelligence

Course Structure: 

Unit 1:  Decision Trees for Classification

• Decision Trees Characteristics
• Gini Impurity Score, Entropy Inpurity Score
• Training Algorithm, Hyperparameters, Computational Complexity, Model Sensitivity and Stability
• First Decision Tree Classification Example
• Detailed Calculation of the Gini Score
• Decision Boundaries, Estimation of Class Probabilities, Making Predictions
• Detailed Calculation of the Entropy Score
• Second Decision Tree Classification Example
• Underfitting, Overfitting, Tree Depth
• Decision Tree Classification Example using the Iris Data Set
• Decision Regions, Confusion Matrix
• Feature Importance, Grid Search

Unit 2: Decision Trees for Regression

• Model Characteristics, Regression Tree Models, Training Algorithm
• Regression Tree Example, Overfitting versus Underfitting
• Regression Tree Example (continues)
• Overfitting
• Model Regularization
• Hyperparameter Optimization via GridSearcCV, Model Regularization via GridSearchCV

Unit 3: Decision Tree Example

• The California Housing Data Set
• Instantiate Regression Tree, Use GridSearchCV, Evaluate Performance, Visualize Tree
• Instantiate Random Forest Regressor, Use GridSearchCV, Evaluate Performance
• Instantiate Random Forest Regressor using Hyperparameters found through GridSearcCV, Evaluate Performance
• Decision Tree Overview (assigned reading)

Unit 4: Code Generation using Large Language Models

• Getting Started with the PaLM API (Google), Text and Code Generation
• A Possible Prompt Structure, Build your Prompt, Generate Code, Run the Generated Code
• A Second Code Generation Trial using a Higher Temperature, Run the Generated Code
• Getting Started with OpenAI API, Code Generation using  DaVinci Model and GPT-4