Scientific research project. Developed with Prof. Dr. João Roberto Bertini Jr.
The final paper was presented in the 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) and can be accessed here.
All the project dependencies can be found under requirements.txt.
pip install -e kaogexp
If you have some problems with the instalation of the kaog module, you can try it manualy following the steps
avaliable in here.
All the experiments and results are located in the main directory. All the scripts used in the paper in the following links:
The KAOGExp can be used creating an object from KAOGExp. To explain one instance, you can use the method explicar(),
that requires some parameters like the method to use (Counterfactual or anything implementing MethodAbstract), the
data cleaner and the normalizer associated with the dataset and model.
More documentation should be added later.