This is a course with focus of learning concepts in machine learning using scikit-learn. This course will build upon prior skills in:
- pandas
- visualising data
- fundamental statistics
- fundamental linear algebra
All lecture codes and exercises can be found in the course Github repo.
This course mostly focuses on supervised learning, both regression and classification, but has small parts in unsupervised learning, dimension reductions and artificial neural networks.
| Week | Content |
|---|---|
| 7 | Regression Linear, gradient descent, sklearn |
| 8 | Polynomial, overfitting, underfitting, regularization, cross-validation |
| 9 | Classification: Logistic regression, KNN |
| 10 | GridsearchCV, SVM, lab1 |
| 11 | Descision tree, Random forest, NLP intro, Naive bayes, lab1 |
| 12 | Unsupervised: K-means, PCA lab1 |
| 13 | ANN intro, Repetition |
| 14 | Repetition, Exam (tuesday) |
| 15 | Apply for LIA, self-study, re-exams on other courses |
Many exercises and lecture materials are in form of Jupyter notebooks with .ipynb extensions. Sometimes GitHub may not load them correctly for preview, then you can use Open in Colab, which is an addon in Chrome to open the notebook in Colab. Alternatively, you can go to jupyter nbviewer, and paste the link to the notebook for previewing. When working with exercises it is important that you create your own notebooks (.ipynb) or script files (.py).