ParamZX
An update to PyZX to support parameterised reduction and GPU evaluations, as described in the paper 'Fast classical simulation of quantum circuits via parametric rewriting in the ZX-calculus'
Install / Use
/learn @mjsutcliffe99/ParamZXREADME
ParamZX
This repository hosts the necessary code to demonstrate the parameterised ZX-calculus reduction (and GPU parallel evaluations) method outlined in the paper 'Fast classical simulation of quantum circuits via parametric rewriting in the ZX-calculus'.
Contained within is:
- an updated version of the PyZX library (https://github.com/Quantomatic/pyzx) to support parameterised diagrams and reduction
- a Jupyter notebook which demonstrates how Clifford+T circuits may be reduced into parameterised scalars (and structured to be GPU-ready)
- CUDA code that reads these GPU-ready parameterised scalars and performs speedy evaluations upon then (and benchmarks the speed measurements)
Attribution
Anyone is welcome to use this work, and for those who do it would be appreciated if you cite the related paper https://arxiv.org/abs/2403.06777:
<pre> @misc{sutcliffe2024fastclassicalsimulationquantum, title={Fast classical simulation of quantum circuits via parametric rewriting in the ZX-calculus}, author={Matthew Sutcliffe and Aleks Kissinger}, year={2024}, eprint={2403.06777}, archivePrefix={arXiv}, primaryClass={quant-ph}, url={https://arxiv.org/abs/2403.06777}, } </pre>