You are here:

Cmotions Python Course

The Python course of the MARUG will be offered in cooperation with Cmotions and will be given in English. During this course, Cmotions will teach you all basics of Python. After the course, you will receive a prove that you have taken part in the course. So a nice opportunity to boost your CV!

With this course we want to give you a kickstart in Python. We’ll show you how basic pre-processing steps are made, like importing data, handling data, how to handle outliers and missing values, and show you how to train- and evaluate models. We’ll start easy, we’ll begin with an explanation of what Python is and get you acquainted with coding. We’ll show you some functions and how to use them. As we proceed, you´re going to have to find out more yourself. In this way, we teach you where and how to look for help, which is essential for effective programming.

Module 1: 14th of February

The first module is an introduction to Python. You’re going to get acquainted with fundamentals of this modelling language. You’ll learn about working in Jupyter Notebook, installing new Libraries/Packages, Variable Types, Indexing and Slicing. After this module, you’ll know the absolute basics of Python.

 Module 2: 21st of February

You will learn all about Data Handling in Python. You start working with Pandas DataFrames, importing CSV-files, (conditional) subsetting, summarizing, aggregating, creating-  renaming- and deleting columns, and learn some basic plotting. After this module, you’ll be able to perform the most important steps in data-processing.

 Module 3: 28th of February

Now you’ve learned how to process data, it’s time to start modelling. We’ll use the most used Machine Learning library in our field: Sci-kit Learn. You will learn how to write your own functions, create data-pipelines that automatically handle missings, and create train- and test- sets. After that, you’ll learn to build three types of model: a Decision Tree, a Logistic Regression and an XgBoost model, and use evaluation metrics (Accuracy, AUC, F1-score) to decide on the best model. If there is time, we will talk about automatic Cross-Validation and GridSearch so you can let Python select the optimal model for you!