Python for NeuroGrads
Join us for a two-day workshop on Python programming on May 3rd + 4th (9-16), 2021 in Faculty Club, the Panum Institute, building 16.6 from 9-16 both days.
Please note that this workshop is only for NeuroGrads.
Applied Data Science with Python
Would you like to be able to predict key figures for your work? With Data Science and Machine Learning algorithms, it is possible to create highly accurate predictive models. You will get a thorough introduction to Data Science and Machine Learning in Python, where focus will be on actual programming in Python and understanding the relevant methods and principles in Data Science. You will learn the basic libraries which are necessary to produce predictive models and you will get an introduction to supervised machine learning models and how these should be interpreted.
This course is for you, who wants to learn how to use Python in your work with Machine Learning. Maybe you are working with data on a daily basis, and you would like to gain more value by using Machine Learning models.
You are expected (but not required) to have a basic understanding of programming in Python or other programming languages. To gain full benefit from the course, concepts such as variables, loops, if-statements and functions should be relatively familiar or rapidly acquired prior to/during the course.
The course consists of a theoretical overview of models and relevant theory, and with a strong focus on hands-on assignments. This way you will quickly and easily be able to use your newly acquired skills in practice.
- Module 1: Tools
Data science requires a series of libraries that you should be familiar with. We will go through Pandas, Scikit-learn and more.
- Module 2: Data and plotting
In real life a lot of time is spent on loading and cleaning data. We will go through procedures for loading data, performing imputation and one-hot encoding.
- Module 3: Supervised learning models
Modelling is essential to make good predictions. A subset of models from scikit-learn will be presented, one of which will be Random Forest.
- Module 4: Model Evaluation
To ensure consistent results on new data, it is important to evaluate a model with the proper techniques. We will use cross-validation to validate the model, and go into tuning of hyperparameters to avoid overfitting.
Mathias Kvist Aarup works as a Data Scientist, where Python is used as the primary programming language. Mathias has a master degree in Mathematical Finance from Aarhus University, where he has also conducted class room teaching for multiple years. He firmly believes that Python is the best language for Data Science at the moment.