This is supplementary code for the paper "A Lightweight Protocol for Matchgate Fidelity Estimation", used to simulate Algorithm 1 as described in the text, using Python 3.11 and qiskit. We recommend using a virtual environment to install the project, however the code can be directly copied from the mgbenchmark folder.
The source code is located in the mgbenchmark folder, which contains the __init__.py, main.py, and utils.py files.
The examples folder contains .csv files with quoted data, and four Jupyter Notebooks:
matchgate_benchmarking.ipynb: A demo of the Matchgate Fidelity Estimation protocol (with brief discussion).matchgate_benchmarking_data_collection.ipynb: A demo of the protocol with different noise models, as well as code used to collect data for the paper.matchgate_tomography.ipynb: A demo of efficient matchgate tomography, first described by Oszmaniec, Dangniam, Morales and Zimboras.data_plots.ipynb: Code used to make the figures.
The project uses toml for configuration instead of setup.py. The configuration file is located in pyproject.toml. We've used Pyright for static type checking, pre-commit for code formatting, Black for code formatting and Ruff for linting. The configuration for these tools is located in the .pre-commit-config.yaml and ruff.yaml file. This was included for convenience, and can be ignored.
To install the project, clone the repository and run (on some systems you may need to use python3 instead of python):
python3 -m venv .venv
source .venv/bin/activate
pip install -U pip setuptools
pip install -r requirements.txt
pre-commit installThen install the project using:
pip install -e .