SD204 - linear models (2016-2018)

Linear models and a bit of sparsity
linear models
sparse models
Left: points cloud is showed (green points), and a line goes through the center of mass of the point cloud. The orthogonal projection of the points are connected by dashes to their projections (in black). Right: the curve (in black) show the associated variance of the projected samples. The current point (in green) represent the variance of the black projected points from the left part.

This is an undergraduate course on linear models.

Slides and Courses

  1. Introduction: LeastSquare_1D_en.pdf, LeastSquare_1D_en.ipynb, LeastSquare_1D_en.html

  2. Least squares : LeastSquare_Def_en.pdf, LeastSquare_1D_fr.ipynb, LeastSquare_Def_en.html

  3. Least squares (properties) : LeastSquare_Prop_fr.pdf

  4. SVD: SVD_en.pdf (in French), SVD.ipynb, SVD.html

  5. PCA: PCA_en.pdf (in French), PCA.ipynb, PCA.html

  6. IntroIC: IntroIC_en.pdf (in French), IntroIC.ipynb, IntroIC.html

  7. Bootstrap: Bootstrap_en.pdf,

  8. IntroTests: IntroTests_en.pdf, forward_variable_selection.ipynb, forward_variable_selection.html

  9. Ridge: Ridge_en.pdf, Ridge_en.ipynb, Ridge_en.html

  10. Lasso: Lasso_en.pdf, Lasso_fr.ipynb, Lasso_fr.html, functions_Lasso.py, prox_collection.py

  11. Coordinate descent: CD_en.pdf (in French), CD_fr.ipynb, CD_fr.html

  12. GLM: GLM_en.pdf (in French), GLM_fr.ipynb, GLM_fr.html

  13. Generalization: Generalization_en.pdf, Generalization_fr.ipynb, Generalization_fr.html

  14. Categorical variables: CategoricalVariables.ipynb, CategoricalVariables.html