qujax

qujax: Simulating quantum circuits with JAX - Published in JOSS (2023)

https://github.com/cqcl/qujax

Science Score: 98.0%

This score indicates how likely this project is to be science-related based on various indicators:

✓
CITATION.cff file
Found CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
✓
DOI references
Found 6 DOI reference(s) in README and JOSS metadata
✓
Academic publication links
Links to: joss.theoj.org
○
Committers with academic emails
○
Institutional organization owner
✓
JOSS paper metadata
Published in Journal of Open Source Software

Keywords

jax python quantum

Keywords from Contributors

mesh

Scientific Fields

Engineering Computer Science - 32% confidence

Last synced: 4 months ago · JSON representation ·

Repository

Simulating quantum circuits with JAX

Basic Info

Host: GitHub
Owner: CQCL
License: apache-2.0
Language: Python
Default Branch: develop
Homepage: https://cqcl.github.io/qujax/
Size: 4.3 MB

Statistics

Stars: 71
Watchers: 5
Forks: 13
Open Issues: 10
Releases: 13

Topics

jax python quantum

Created over 3 years ago · Last pushed 4 months ago

Metadata Files

Readme License Citation

qujax

qujax is a JAX-based Python library for the classical simulation of quantum circuits. It is designed to be simple, fast and flexible.

It follows a functional programming design by translating circuits into pure functions. This allows qujax to seamlessly interface with JAX, enabling direct access to its powerful automatic differentiation tools, just-in-time compiler, vectorization capabilities, GPU/TPU integration and growing ecosystem of packages.

qujax can be used both for pure and for mixed quantum state simulation. It not only supports the standard gate set, but also allows user-defined custom operations, including general quantum channels, enabling the user to e.g. model device noise and errors.

A summary of the core functionalities of qujax can be found in the Quick start section. More advanced use-cases, including the training of parameterised quantum circuits, can be found in the Examples section of the documentation.

Installation

qujax is hosted on PyPI and can be installed via the pip package manager pip install qujax

Quick start

Important note: qujax circuit parameters are expressed in units of $\pi$ (e.g. in the range $[0,2]$ as opposed to $[0, 2\pi]$).

Start by defining the quantum gates making up the circuit, the qubits that they act on, and the indices of the parameters for each gate.

A list of all gates can be found here (custom operations can be included by passing an array or function instead of a string).

```python from jax import numpy as jnp import qujax

List of quantum gates

circuit_gates = ['H', 'Ry', 'CZ']

Indices of qubits the gates will be applied to

circuitqubitinds = [[0], [0], [0, 1]]

Indices of parameters each parameterised gate will use

circuitparamsinds = [[], [0], []]

qujax.printcircuit(circuitgates, circuitqubitinds, circuitparamsinds);

q0: -----H-----Ry[0]-----◯---

|

q1: ---------------------CZ--

```

Translate the circuit to a pure function param_to_st that takes a set of parameters and an (optional) initial quantum state as its input.

```python paramtost = qujax.getparamstostatetensorfunc(circuitgates, circuitqubitinds, circuitparams_inds)

paramtost(jnp.array([0.1]))

Array([[0.58778524+0.j, 0. +0.j],

[0.80901706+0.j, 0. +0.j]], dtype=complex64)

```

The optional initial state can be passed to param_to_st using the statetensor_in argument. When it is not provided, the initial state defaults to $\ket{0...0}$.

Map the state to an expectation value by defining an observable using lists of Pauli matrices, the qubits they act on, and the associated coefficients.

python st_to_expectation = qujax.get_statetensor_to_expectation_func([['Z']], [[0]], [1.])

Combining param_to_st and st_to_expectation gives us a parameter to expectation function that can be automatically differentiated using JAX.

```python from jax import valueandgrad

paramtoexpectation = lambda param: sttoexpectation(paramtost(param)) expectationandgrad = valueandgrad(paramtoexpectation) expectationandgrad(jnp.array([0.1]))

(Array(-0.3090171, dtype=float32),

Array([-2.987832], dtype=float32))

```

Mixed state simulations are analogous to the above, but with calls to get_params_to_densitytensor_func and get_densitytensor_to_expectation_func instead.

A more in-depth version of the above can be found in the Getting started section of the documentation. More advanced use-cases, including the training of parameterised quantum circuits, can be found in the Examples section of the documentation.

Converting from TKET

A pytket circuit can be directly converted using the tk_to_qujax and tk_to_qujax_symbolic functions in the pytket-qujax extension. See pytket-qujax_heisenberg_vqe.ipynb for an example.

Contributing

You can open a bug report or a feature request by creating a new issue on GitHub.

Pull requests are welcome! To open a new one, please go through the following steps:

First fork the repo and create your branch from develop.
Commit your code and tests.
Update the documentation, if required.
Check the code lints (run black . --check and pylint */).
Issue a pull request into the develop branch.

New commits on develop will be merged into main in the next release.

Citing qujax

If you have used qujax in your code or research, we kindly ask that you cite it. You can use the following BibTeX entry for this:

bibtex @article{qujax2023, author = {Duffield, Samuel and Matos, Gabriel and Johannsen, Melf}, doi = {10.21105/joss.05504}, journal = {Journal of Open Source Software}, month = sep, number = {89}, pages = {5504}, title = {{qujax: Simulating quantum circuits with JAX}}, url = {https://joss.theoj.org/papers/10.21105/joss.05504}, volume = {8}, year = {2023} }

Owner

Name: Cambridge Quantum
Login: CQCL
Kind: organization
Location: Cambridge, UK

Website: http://www.cambridgequantum.com
Repositories: 48
Profile: https://github.com/CQCL

Quantum Software and Technologies

JOSS Publication

qujax: Simulating quantum circuits with JAX

Published

September 12, 2023

DOI

10.21105/joss.05504

Volume 8, Issue 89, Page 5504

Authors

Samuel Duffield

Quantinuum

Gabriel Matos

Quantinuum, University of Leeds

Melf Johannsen
Quantinuum

Editor

Lucy Whalley

Citation (CITATION.cff)

cff-version: "1.2.0"
authors:
- family-names: Duffield
  given-names: Samuel
  orcid: "https://orcid.org/0000-0002-8656-8734"
- family-names: Matos
  given-names: Gabriel
  orcid: "https://orcid.org/0000-0002-3373-0128"
- family-names: Johannsen
  given-names: Melf
contact:
- family-names: Duffield
  given-names: Samuel
  orcid: "https://orcid.org/0000-0002-8656-8734"
doi: 10.5281/zenodo.8268973
message: If you use this software, please cite our article in the
  Journal of Open Source Software.
preferred-citation:
  authors:
  - family-names: Duffield
    given-names: Samuel
    orcid: "https://orcid.org/0000-0002-8656-8734"
  - family-names: Matos
    given-names: Gabriel
    orcid: "https://orcid.org/0000-0002-3373-0128"
  - family-names: Johannsen
    given-names: Melf
  date-published: 2023-09-12
  doi: 10.21105/joss.05504
  issn: 2475-9066
  issue: 89
  journal: Journal of Open Source Software
  publisher:
    name: Open Journals
  start: 5504
  title: "qujax: Simulating quantum circuits with JAX"
  type: article
  url: "https://joss.theoj.org/papers/10.21105/joss.05504"
  volume: 8
title: "qujax: Simulating quantum circuits with JAX"

GitHub Events

Total

Watch event: 8
Issue comment event: 1
Push event: 1
Pull request review event: 1
Pull request event: 6
Fork event: 3
Create event: 4

Last Year

Watch event: 8
Issue comment event: 1
Push event: 1
Pull request review event: 1
Pull request event: 6
Fork event: 3
Create event: 4

Committers

Last synced: 5 months ago

All Time

Total Commits: 205
Total Committers: 6
Avg Commits per committer: 34.167
Development Distribution Score (DDS): 0.268

Past Year

Commits: 2
Committers: 2
Avg Commits per committer: 1.0
Development Distribution Score (DDS): 0.5

Top Committers

Name	Email	Commits
Sam Duffield	s@m****m	150
Gabriel Matos	g**s@q**m	28
melf	c**c@m**e	15
dependabot[bot]	4****]	10
Oliver Backhouse	o**e@g**m	1
Enrico Rinaldi	e**i@q**m	1

Committer Domains (Top 20 + Academic)

quantinuum.com: 2 melf.de: 1 mduffield.com: 1

Issues and Pull Requests

Last synced: 4 months ago

All Time

Total issues: 16
Total pull requests: 109
Average time to close issues: about 2 months
Average time to close pull requests: 3 days
Total issue authors: 3
Total pull request authors: 7
Average comments per issue: 0.94
Average comments per pull request: 0.28
Merged pull requests: 102
Bot issues: 0
Bot pull requests: 10

Past Year

Issues: 0
Pull requests: 3
Average time to close issues: N/A
Average time to close pull requests: 9 days
Issue authors: 0
Pull request authors: 2
Average comments per issue: 0
Average comments per pull request: 0.33
Merged pull requests: 1
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

SamDuffield (10)
gamatos (5)
tlaakkonen (1)

Pull Request Authors

SamDuffield (56)
gamatos (30)
dependabot[bot] (14)
cqc-melf (9)
obackhouse (2)
lucydot (2)
erinaldiq (1)

Top Labels

Issue Labels

good first issue (9) enhancement (7) documentation (7) bug (1)

Pull Request Labels

dependencies (14)

Packages

Total packages: 1
Total downloads:
- pypi 2,252 last-month

Total dependent packages: 1
Total dependent repositories: 1
Total versions: 27
Total maintainers: 1

pypi.org: qujax

Simulating quantum circuits with JAX

Homepage: https://github.com/CQCL/qujax
Documentation: https://qujax.readthedocs.io/
License: Apache 2
Latest release: 1.1.0
published about 2 years ago

Versions: 27
Dependent Packages: 1
Dependent Repositories: 1
Downloads: 2,252 Last month

Rankings

Dependent packages count: 4.7%

Stargazers count: 10.2%

Downloads: 10.4%

Average: 11.9%

Forks count: 12.6%

Dependent repos count: 21.8%

Maintainers (1)

samduffield

Last synced: 4 months ago

Dependencies

.github/workflows/docs/requirements.txt pypi

sphinx *
sphinx_rtd_theme *

docs/requirements.txt pypi

jax *
jaxlib *
sphinx >=3
sphinx-autodoc-typehints *
sphinx-rtd-theme >=1.0.0

setup.py pypi

jax *

.github/workflows/build_and_test.yml actions

actions/checkout v3 composite
actions/download-artifact v3 composite
actions/setup-python v4 composite
actions/upload-artifact v3 composite
pypa/gh-action-pypi-publish release/v1 composite

.github/workflows/docs.yml actions

actions/checkout v3 composite
actions/setup-python v4 composite
actions/upload-pages-artifact v1 composite

.github/workflows/lint.yml actions

actions/checkout v3 composite
actions/setup-python v4 composite

qujax

Science Score: 98.0%

Keywords

Keywords from Contributors

Scientific Fields

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

qujax

Installation

Quick start

List of quantum gates

Indices of qubits the gates will be applied to

Indices of parameters each parameterised gate will use

q0: -----H-----Ry[0]-----◯---

|

q1: ---------------------CZ--

Array([[0.58778524+0.j, 0. +0.j],

[0.80901706+0.j, 0. +0.j]], dtype=complex64)

(Array(-0.3090171, dtype=float32),

Array([-2.987832], dtype=float32))

Converting from TKET

Contributing

Citing qujax

Owner

JOSS Publication

qujax: Simulating quantum circuits with JAX

Authors

Editor

Tags

Citation (CITATION.cff)

GitHub Events

Total

Last Year

Committers

All Time

Past Year

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

pypi.org: qujax

Rankings

Maintainers (1)

Dependencies