Based on [ExplainDT](https://github.com/DiegoEmilio01/A-Symbolic-Language-for-Interpreting-Decision-Trees) by Arenas et al.
Reminder: I do not own this implementation. All the credit goes to Arenas et al. [here](https://github.com/DiegoEmilio01/A-Symbolic-Language-for-Interpreting-Decision-Trees). Bünyamin Dincer has only extended the implementation. Everything that is now mentioned additionally was implemented by Bünyamin Dincer based on the implementation by Arenas et al.
---
To test the implementation yourself, we recommend setting everything up first. To do so, we recommend cloning this repository, and following along the ReadMe [here](https://github.com/DiegoEmilio01/A-Symbolic-Language-for-Interpreting-Decision-Trees). After setting everything up, to find out more about our contribution feel free to come back here to find out more.
---
## Our Contribution
- Added a [GUI](GUI_byBuni.py) for the REPL interpreter allowing the extension of plugins
- Added a special drawing panel for the MNIST dataset for testing purposes
- Prototype query builder plugin added.
- Added second [Mushroom](https://archive.ics.uci.edu/dataset/73/mushroom) dataset for testing purposes.
- Updated the REPL Interpreter: If class names have spaces, then one can write "Class A", instead of Class A to prevent errors. Also added generate n command to generate a random instance, and a random partial instance of dimension n
---
To run the GUI for the REPL interpreter simply run the following command on the project's root directory:
```sh
python3 GUI_byBuni.py
```
---
The rest of our work is ran in the same way as in the ReadMe [here](https://github.com/DiegoEmilio01/A-Symbolic-Language-for-Interpreting-Decision-Trees).
