Machine Learning Course 2018

The aim of the course is to introduce students to machine learning. Upon passing it, the participants will be able to apply and understand the theory and implementation principles behind most of the commonly utilized machine learning techniques. The following topics will be covered:

  • Machine Learning (ML) basics
  • Basic classification methods
  • Probabilistic methods in ML
  • Clustering methods
  • Decision trees methods
  • Support Vector Machine
  • Neural Networks
  • Ensemble methods
  • Dimensionality reduction methods
  • Basics of Natural Language Processing



Environment setup

To start working with Python and jupyter notebooks in the computer labs you have to log in to a Linux account. Then execute the command

        source /app/Python/3.6.2/VE/DataScience/bin/activate
This will setup a virtual Python environment with necessary packages. The jupyter notebook is started with command
        jupyter notebook

This should open a browser or a new tab in browser. You can now choose new from the menu to create a new notebook or open an existing notebook.

Installation of the environment on your own machine

We suggest that you use Anaconda or miniconda environment for managing python and its modules. To acomplish this please go to miniconda or Anaconda download sites na follow the instructions there. You should install Python 3.6 !

After you have downloaded and installed any of those environments you should create a Python virtual environment for Data Science. This is achieved by using conda command. This is done by invoking

    	conda create --name datascience python=3.6 numpy scipy  pandas scikit-learn matplotlib jupyter notebook

This will created the environment and install the required packages.

After creating the environment you can activate it using the command

    	source activate datascience

and deactivate usig the command

    	source deactivate datascience

The detailed documentation can be found conda user guide.

After activating the environment please go to the directory where the notebook is located and start the jupyter notebook with the command:

    	jupyter notebook

This should start server and open the browser, or new tab in already open browser with a list of files. Clicking on the notebook file should open it.