Module 3 • Machine Learning with Python – ICARUS Education

Home » Courses » Module 3 • Machine Learning with Python

Course Content

Week 1 • Theoretical background of Machine Learning and K-Nearest Neighbors (KNN)

W1L1 Machine Learning background P1 Supervised, Unsupervised, Semisupervised, Reinforcement learning

W1L1 Machine Learning background P2 Regression vs Classification

W1L1 Machine Learning background P3 Train-Validation-Test splits

W1L1 Machine Learning background P4 Loss functions

W1L1 Machine Learning background P5 Bias-Variance Tradeoff (Overfitting, Underfitting)

W1L1 Machine Learning background P6 Cross-validation

W1L2 KNN (first non-parametric algorithm) P1 KNN classification

W1L2 KNN (first non-parametric algorithm) P2 KNN regression

W1L2 KNN (first non-parametric algorithm) P3 KNN package

W1L2 KNN (first non-parametric algorithm) P4 Choosing k via cross-validation

W1L2 KNN (first non-parametric algorithm) P5 Comparison with logistic regression or MLR

W1L2 KNN (first non-parametric algorithm) P6 Confusion matrices, Accuracy

Week 2 • Regularization

W2L1 Theory P1 Bias-variance tradeoff

W2L1 Theory P2 Regularization controlling variance

W2L1 Theory P3 Loss function modification

W2L1 Theory P4 Types of regularization (l1, l2, mixed)

W2L1 Theory P5 Ridge regression

W2L2 LASSO, Elastic-Net regressions P1 Ridge vs LASSO

W2L2 LASSO, Elastic-Net regressions P2 Combination of Ridge and LASSO

W2L2 LASSO, Elastic-Net regressions P3 Choosing the regularization parameter

1 of 2

About Instructor

Nijat Aliyev

4 Courses