Decision Trees

This work has two components:

• Required is an implementation for sk-learn for a way to handle categorical data (and also mixed data, i.e. categorical and numerical). This implementation is hosted at this fork. Error rate, information gain and gini index are provided in each result tree.

• Then, the solution is applied to the data sets provided.

• To visualise the results, the output of the tree is reported in a graphical output.

The numeric parts of the user's matriculation number of the FH Technikum is used for deciding about the random seed for splitting your data into training & test set.

Installation

At least for macOS environment:

export CC=/usr/bin/clang        
export CXX=/usr/bin/clang++
export CPPFLAGS="$CPPFLAGS -Xpreprocessor -fopenmp"
export CFLAGS="$CFLAGS -I/usr/local/opt/libomp/include"
export CXXFLAGS="$CXXFLAGS -I/usr/local/opt/libomp/include"
export LDFLAGS="$LDFLAGS -L/usr/local/opt/libomp/lib -lomp"
export DYLD_LIBRARY_PATH=/usr/local/opt/libomp/lib

$  pip3 install -r requirements.txt

For dot files install graphviz:

Linux:

$  sudo apt-get install graphviz

MacOS: https://brewinstall.org/install-graphviz-on-mac-with-brew/

Running

$  python3 Exercise2.py