The decision tree classifier is written in Python.
Our code for the projects consists of only one jupyter notebook file.
The notebook file works as a program in a whole and each cell needs to be excuted in the given order.
It also contains all the necessary output and each cell is marked with its functionality.
It is important to have all the necessary modules for the execution downloaded and ready.
We used:
pandas
numpy
matplotlib
sklearn
In this first project for the Introduction to Machine Learning course at UiB we needed to implement a decision tree classifier on continuos data from scratch.
The implementation includes the option for two different impurity measures and the use of pruning.
See project1.pdf for more information on the task.