• Simon Tannert
• Touhidul Alam

The repository is structured as follows:

• code/ contains our implementations for the different tasks in the class. The code is implemented in Python3 (tested on >=3.5)
• results/ contains the output of our implementations
• documentation/ will contain our presentation slides and report

code/ is the root of a Python package. There are subdirectories/subpackages for the different implementation tasks.

• code/lib/reader.py
• code/lib/feature_extraction.py
• code/lib/perceptron.py
• code/lib/naive_bayes.py

To run our programs, you should use the shell scripts provided in the root of our git-repository.

We use docopt as command line parser, you can install it with the following command

python3 -m pip install --user docopt