README.md

Quantum Seeded Synthesis BQSKit Pass

A Python package that implements machine learning powered seeded quantum circuit synthesis.

Installation

This is available for Python 3.8+ on Linux, macOS, and Windows.

git clone [email protected]:mtweiden/qseed.git
cd qseed
pip install -e .

Basic Usage

The easiest way to run QSeed is to call the run.py script. It requires a qasm file as input. Adding the --qsearch flag will perform synthesis with QSearch.

python run.py qasm/<qasm_family>/mapped-<qasm_file> [--qsearch]

Outputs are stored in the compiled_circuits directory.

How to Cite

To cite qseed:

@misc{
    weiden2023improving,
    title={Improving Quantum Circuit Synthesis with Machine Learning}, 
    author={Mathias Weiden and Ed Younis and Justin Kalloor and John Kubiatowicz and Costin Iancu},
    year={2023},
    eprint={2306.05622},
    archivePrefix={arXiv},
    primaryClass={quant-ph}
}

To cite BQSKit, you can use the software disclosure.