How gardkl/project1 works
This coursework project runs a notebook-based study of polynomial regression on Runge’s function. Its documented workflow covers OLS, Ridge and Lasso, bootstrap bias–variance analysis, cross-validation and gradient-based optimization; figures are reproducible notebook outputs. The sampled regression module implements data generation, polynomial features, OLS/Ridge fitting and prediction, and metrics. The notebook was not sampled, so its internal call structure and links to library functions remain unverified.
Main entry points: Project1_gard.ipynb Project notebook · regression.py Fit and predict · results Result figures
Components 16
Notebook workflow
- Project notebook
Project1/code/Project1_gard.ipynb - Result figures
Project1/results
Regression library
- Runge data
Project1/code/regression.py - Polynomial features
Project1/code/regression.py - OLS fitting
Project1/code/regression.py - Ridge fitting
Project1/code/regression.py - Fit and predict
Project1/code/regression.py - Model metrics
Project1/code/regression.py - Lasso regression
Project1/code/Project1_gard.ipynb
Resampling and selection
- Bootstrap analysis
Project1/code/Project1_gard.ipynb - K-fold selection
Project1/code/Project1_gard.ipynb
Gradient optimization
- Plain gradient descent
Project1/code/Project1_gard.ipynb - Momentum descent
Project1/code/Project1_gard.ipynb - Adaptive optimizers
Project1/code/Project1_gard.ipynb - Stochastic descent
Project1/code/Project1_gard.ipynb
Other components
- Coursework user
Connections 15
- Coursework user → Project notebook: runs cells
README.md - Project notebook → Result figures: reproduces figures
README.md - Project notebook → Runge data: uses library
- Project notebook → Lasso regression: studies
README.md - Project notebook → Bootstrap analysis: analyzes
README.md - Project notebook → K-fold selection: selects models
README.md - Project notebook → Plain gradient descent: optimizes with
README.md - Project notebook → Momentum descent: optimizes with
README.md - Project notebook → Adaptive optimizers: optimizes with
README.md - Project notebook → Stochastic descent: optimizes with
README.md - Runge data → Polynomial features: generates samples for
Project1/code/regression.py - Fit and predict → Polynomial features: builds
Project1/code/regression.py - Fit and predict → OLS fitting: uses when unregularized
Project1/code/regression.py - Fit and predict → Ridge fitting: uses when regularized
Project1/code/regression.py - Fit and predict → Model metrics: returns predictions for evaluation
Project1/code/regression.py