Data

In order to work, the pipeline needs the evaluation data. While you can change the location of the raw_data, clean_data and outputs folder as described in lib. You can't change the structure of subfolders.

raw_data

All the evaluation data that hasn't been cleaned, should be found here. If you didn't change anything in the constants.py file, the underlying structure should always be:

├── Bioinformatik
│   └── Bioinformatik_SoSe2019
│       ├── daten.csv
│       ├── kommentare.csv
│       ├── mittelwerte.csv
│       └── summen.csv
├── Informatik
│   └── Informatik_SoSe2019
│       ├── daten.csv
│       ├── kommentare.csv
│       ├── mittelwerte.csv
│       └── summen.csv
├── Mathematik
│   └── Mathematik_SoSe2019
│       ├── daten.csv
│       ├── kommentare.csv
│       ├── mittelwerte.csv
│       └── summen.csv
└── participation
    └── SoSe2019.csv

Note that the provides summer term simply is an example, you can have any semester, as long as the folder names start with the institute.

clean_data

The folder structure of the clean data will be generated by cleanup.ipynb, if you want to change it, you can do so by rewriting the last two code cells in the notebook. Keep in mind you should also then change the constants to reflect your new layout.

If left untouched, the generated structure will be similar to the one in raw_data

├── Bioinformatik
│   └── WiSe2022-23
│       ├── LeKo.csv
│       ├── S.csv
│       ├── Ü.csv
│       ├── VL.csv
│       └── all.csv
├── Informatik
│   └── WiSe2022-23
│       ├── LeKo.csv
│       ├── S.csv
│       ├── Ü.csv
│       ├── VL.csv
│       └── all.csv
├── Mathematik
│   └── WiSe2022-23
│       ├── LeKo.csv
│       ├── S.csv
│       ├── Ü.csv
│       ├── VL.csv
│       └── all.csv
└── participation
    └── WiSe2022-23.csv

outputs

This folder is only used by the old-report.ipynb. Similar to clean_data the result structure can be changed by changing the last code cell and the constants. If left untouched the structure will be

├── WiSe2022-23.md
└── pdfs
    └── WiSe2022-23.pdf

Note that the pipeline only generated the markdown files, which you in turn can compile into PDF files by using pandoc for example. Simply use pandoc -o pdfs/output.pdf input.md and the generated Markdown table will become a PDF.