The referenceqvm
is the reference implementation of the QVM outlined in the
arXiv:1608:03355 by Robert Smith, Spike Curtis, and Will Zeng. It is a research package that
supports rapid prototyping and development of quantum programs using pyQuil.
Currently, this QVM supports a subset of functionality in the Quil specifications, excepting certain functions (DEFCIRCUIT, WAIT, NOP).
Noise models (dephasing, Kraus operators), parametrization with bits in classical memory, and other features will be added in future releases.
You can install reference-qvm directly from the Python package manager pip
using:
pip install referenceqvm
To instead install reference-qvm from source, clone this repository, cd
into it, and run:
pip install -r requirements.txt -e .
This will install the reference-qvm's dependencies if you do not already have them.
We use tox and pytest for testing. Tests can be executed from the top-level directory by simply running:
tox
The setup is currently testing Python 2.7 and Python 3.6.
We use sphinx to build the documentation. To do this, navigate into pyQuil's top-level directory and run:
sphinx-build -b html ./docs/source ./docs/build
To view the docs navigate to the newly-created docs/build
directory and open
the index.html
file in a browser. Note that we use the Read the Docs theme for
our documentation, so this may need to be installed using pip install sphinx_rtd_theme
.
The qvm can be accessed in a similar way to the Forest QVM access.
Start by importing the synchronous connection object from the referenceqvm.api
module
from referenceqvm.api import SyncConnection
and initialize a connection to the reference-qvm
qvm = SyncConnection()
By default, the Connection object uses the wavefunction transition type.
Then call the qvm.wavefunction(prog)
method to get back the classical memory and the
pyquil.Wavefunction object given a pyquil.quil.Program object prog
.
The reference-qvm has the same functionality as Forest QVM and is useful for testing
small quantum programs on a local machine. For example, the same code (up to the
referenceqvm.api
import) can be used to simulate pyquil programs.
>>> import pyquil.quil as pq
>>> import referenceqvm.api as api
>>> from pyquil.gates import *
>>> qvm = api.SyncConnection()
>>> p = pq.Program(H(0), CNOT(0,1))
<pyquil.pyquil.Program object at 0x101ebfb50>
>>> qvm.wavefunction(p)[0]
[(0.7071067811865475+0j), 0j, 0j, (0.7071067811865475+0j)]
SyncConnection can also initialize a QVM that does not return a wavefunction but instead a unitary corresponding to the pyquil program. This can be extremely useful in terms of debugging and understanding gate physics. For example, we can examine the unitary for a CNOT operator.
>>> import pyquil.quil as pq
>>> import referenceqvm.api as api
>>> from pyquil.gates import CNOT
>>> qvm = api.SyncConnection(type_trans='unitary')
>>> p = pq.Program(CNOT(1, 0))
>>> u = qvm.unitary(p)
>>> print(u)
[[ 1.+0.j 0.+0.j 0.+0.j 0.+0.j]
[ 0.+0.j 1.+0.j 0.+0.j 0.+0.j]
[ 0.+0.j 0.+0.j 0.+0.j 1.+0.j]
[ 0.+0.j 0.+0.j 1.+0.j 0.+0.j]]
If you use the reference-qvm please cite the repository as follows:
bibTex:
@misc{rqvm2017.0.0.1,
author = {Rigetti Computing",
title = {Reference-QVM},
year = {2017},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/rigetticomputing},
commit = {the commit you used}
}
and the paper outlining the Mathematical specification of the quantum-abstract-machine:
bibTeX:
@misc{1608.03355,
title={A Practical Quantum Instruction Set Architecture},
author={Smith, Robert S and Curtis, Michael J and Zeng, William J},
journal={arXiv preprint arXiv:1608.03355},
year={2016}
}