Skip to content

Jupyter notebooks for the Fundamentals of Machine Learning course that is held at FEI, VSB-TU Ostrava

Notifications You must be signed in to change notification settings

rasvob/VSB-FEI-Fundamentals-of-Machine-Learning-Exercises

Repository files navigation

Fundamentals of Machine Learning

The course is held at FEI, VSB-TU Ostrava

Course information may be found here.

You can find more details about the course at my homel.

Feel free to contact me (radek.svoboda@vsb.cz) if you have any questions or want to discuss any topic from the course 😊

All authorship is mentioned where possible.

📊 Exercises

Exercise 1

The aim of the exercise is to get an overview of the basic capabilities of the Pandas, Matplotlib and Seaborn libraries and be able to setup a Python Virtual Enviroment (venv)

Jupyter Notebook

Google Colab

Exercise 2

The aim of the exercise is to learn basic techniques for visualization creation and interpretation using Matplotlib and Seaborn libraries.

Jupyter Notebook

Google Colab

Exercise 3

Goal of the excercise is to learn about more advanced vizualization techniques using Matplotlib and Seaborn libraries.

Jupyter Notebook

Google Colab

Exercise 4

Goal of the excercise is to learn how to use K-means implementation in the Scikit-learn library to perform clustering and subsequent cluster analysis on a Titanic dataset.

Jupyter Notebook

Google Colab

Exercise 5

We will learn how to use another clustering algorithm - Hierarchical (or Agglomerative) clustering. The base principles and important hyper-parameters will be explained.

Jupyter Notebook

Google Colab

Exercise 6

The goal of this excercise is to complete the hands-on experience task with similar task description as the first project has.

Jupyter Notebook

Google Colab

Exercise 7

Goal of the excercise is to code selected part of the Decision tree algorithm which is focused on the optimum split part using gini index.

After that the scikit-learn implementation of the Decision tree basic usage will be demonstrated.

Jupyter Notebook

Google Colab

Exercise 8

Goal of the excercise is to learn how to use Scikit-learn library for a classification tasks and evaluate the performance of the proposed models.

Jupyter Notebook

Google Colab

Exercise 9

Goal of this excercise is to complete the hands-on experience of the classification task.

Jupyter Notebook

Google Colab

Exercise 10

Goal of the excercise is to learn how to use Scikit-learn library for a regression tasks employing various linear regression models and moreover evaluate the performance of the proposed models.

Jupyter Notebook

Google Colab

Exercise 11

Goal of the excercise is to learn how to use basic deep learning models in Scikit-learn and Keras.

Jupyter Notebook

Google Colab

Exercise 12

Goal of the excercise is to learn how to save trained models and use selected advanced libraries like Plotly or Optuna.

Jupyter Notebook

Google Colab

Notebooks from the lectures

Credit goes to prof. Ing. Jan Platoš, Ph.D.

Sloths classification

Jupyter Notebook

Google Colab

Diamond price prediction

Jupyter Notebook

Google Colab

💡 Notes

How to create a Python Virtual Enviroment named venv

Create venv

python -m venv venv

Activate venv

  • Activate venv in Windows
.\venv\Scripts\Activate.ps1
  • Activate venv in Linux
source venv/bin/activate

Intall python packages

pip install jupyter "jupyterlab>=3" "ipywidgets>=7.6"
pip install pandas matplotlib requests seaborn scipy scikit-learn optuna tensorflow plotly==5.18.0

🚀 Run Jupyter lab

jupyter lab