QUACK

Implementation of the Quantum Aligned Centroid Kernel algorithm. This project is not maintained. It has been published as part of the following paper: https://arxiv.org/abs/2405.00304

Installation

All required packages should automatically be installed by creating a docker container from the provided Dockerfile.

Usage

• An example of how to use QUACK is given in train_save_load_model.py.
• The parameters and results of the models used in the paper can be found in models/models_test2.paper
• Due to legacy reasons, the validation set is called "test" and the test set "test2"

Datasets

• census_income (CC BY 4.0 license, https://archive.ics.uci.edu/dataset/20/census+income) is provided in datasets/preprocessed, for the other datasets see datasets/datasets.md
• If you want to use your own datasets, overwrite the load_dataset function in datasets.py.