This repository holds the work and results of the Mini Challenge in the module Applied Machine Learning (aml) at FHNW.

• data/: The data directory holds all .csv files from the PKDD'99 Financial Data Set.
  • translation_mappings.json: The translation_mappings.json holds all variable mappings from Czech to English. It is utilized by the DataLoader and translates the variables into english at the onloading.
• images/: Inside the images folder the Entity Relationship Diagram of the Data Set can be found.
• src/: The source directory contains outsourced code such as utility functions and classes.
  • data_utils.py: The data_utils.py file contains the DataLoader class that streamlines the process of loading the dataset into memory.
  • plot_utils.py: The plot_utils.py file holds all plotting functions that are used during exploration and analysis inside the main notebook.ipynb.
  • train_utils.py: The train_utils.py file houses utility functions used for training like the cross_validate function - The centerpiece for training the explored models.
• notebook.html and notebook.ipynb: The main notebooks contain all required specifications from data preparation to model explanations.
• USE-OF-AI.md: This markdown notes how assistants like ChatGPT and GitHub CoPilot were used.
• README.md: This markdown.
• requirements.txt: This textfile contains all dependencies used in the project.