Python
PythonΒΆ
In this class, we will use Python in order to do simulations, illustrations, implementation and test of models etc. We do not require any programming backround in Python, but at least some experience in any programming language would be helpful.
The most important Python libraries for our purposes will be:
numpy (numerical computing library; mostly to do matrix and vector calculations)
pandas (library for data analysis and manipulation, particularly useful for real world data)
scikit-learn (machine learning library)
scipy (scientific computing library; contains many useful tools such as optimization and integration algorithms, special functions etc.)
matplotlib (plotting library)
tensorflow (fast machine learning library with focus on deep learning)
gpflow (machine learning library which uses tensorflow for Gaussian processes)
Feel free to discuss any problems with us regarding Python and the content of the lecture. During the exercise classes, you will have time to practice your programming skills and to understand the theoretical backround.
On the official Python webpage, you can find some beginners guide. Moreover, scikit-learn is a very beginner friendly machine learning library which also offers useful tutorials.
To keep things simple, we decided to use Jupyter Notebooks for coding which are very easy to use and easy to read. Moreover, we will use the cloud service Google Colab to share and to execute these notebooks. In this way, you do not necessarily need to have a Python installation on your computer. Only a webbrowser is required. If you do want a local installation anaconda comes with handy tools like spyter and jupyter preinstalled.