v0.41.0

(Full Changelog)

Highlights

📓 The Layerwise Richardson Extrapolation (LRE) user guide is complete! The user guide contains information about both the ins and outs of using the implementation, as well as covering the theory behind the technique so you can make judgements about when to apply the technique. In addition to finishing the user guide, we also have a new tutorial comparing both the performance and overhead needed for LRE and ZNE. Big thanks to @purva-thakre and @FarLab for the documentation!

📹 As part of launching LRE we made a short tutorial video to showcase the technique, along with how to use it. Check it out here!

🧑‍🔬 First time contributor @jpacold recreated results from a paper on phase transitions in the Ising model. Both the paper authors, and @jpacold both used Mitiq's ZNE module to apply error mitigation. This is both an informative tutorial on turning physics problems into something amenable on a quantum computer, and a class in applying error-mitigation.

✨ Enhancements

• Ensure LRE compatibility with all supported frontends (#2547) [@natestemen]

🧑🏽‍💻 Developer Improvements

• remove failing test; simplify layerwise ZNE tests (#2545) [@natestemen]

📞 Call for ideas

We're currently looking into what features we could add to make Mitiq more noise-aware. If you have ideas and features requests in this area, do make a post on the GitHub discussion here!

v0.40.0

(Full Changelog)

Highlights

🔉 A new quantum error-mitigation technique is available in Mitiq! Layerwise Richardson Extrapolation is now available through mitiq.lre.execute_with_lre. Documentation is also available in the user guide, with more advanced docs and demonstrations coming in the next release. Special thanks to Purva Thakre for this contribution!

🥇 We had two first time contributions from @ecarlander and @mbrotos! Thank you both for your contributions!

🛡️ A helpful error message is raised when passing data of the incorrect type to the MeasurementResult class, where before it silently gave confusing results.

✨ Enhancements

• LRE Executors (#2499) [@purva-thakre]
• LRE Inference Functions (#2447) [@purva-thakre]
• Raise TypeError when a dictionary is passed to MeasurementResult constructor (#2523) [@natestemen]

📓 Documentation

• Add theory, intro and use case pages of LRE user guide (#2522) [@purva-thakre]
• add QOSS survey banner (#2533) [@natestemen]
• Update broken link in the readme (#2528) [@purva-thakre]
• Move class documentation to class docstrings (#2525) [@natestemen]
• QSE docs cleanup (#2490) [@natestemen]
• Add tags to tutorials (#2467) [@purva-thakre]
• Correct CDR and VNCDR acronyms in example (#2479) [@bdg221]
• added roadmap link to readme (#2468) [@ecarlander]
• Update ibmq-backends.md (#2474) [@mbrotos]

🧑🏽‍💻 Developer Improvements

• Remove make requirements (#2481) [@purva-thakre]
• Fix flaky REM test / refactor (#2464) [@natestemen]

📦 Dependency Updates

• Bump pyscf from 2.6.2 to 2.7.0 (#2518) [@dependabot]
• Bump pyqrack from 1.30.24 to 1.30.30 (#2521) [@dependabot]
• Bump pyqrack from 1.30.22 to 1.30.24 (#2497) [@dependabot]
• Update qiskit requirement from ~=1.2.1 to ~=1.2.2 (#2507) [@dependabot]
• Bump qibo from 0.2.10 to 0.2.12 (#2506) [@dependabot]
• Update qiskit-aer requirement from ~=0.15.0 to ~=0.15.1 (#2504) [@dependabot]
• Update qiskit requirement from ~=1.2.0 to ~=1.2.1 (#2503) [@dependabot]
• Bump pyqrack from 1.30.20 to 1.30.22 (#2489) [@dependabot]
• Update qiskit-aer requirement from ~=0.14.2 to ~=0.15.0 (#2484) [@dependabot]
• Bump pyqrack from 1.30.8 to 1.30.20 (#2487) [@dependabot]
• Update cirq-core requirement from <1.4.0,>=1.0.0 to >=1.0.0,<1.5.0 (#2390) [@dependabot]
• Update qiskit requirement from ~=1.1.1 to ~=1.2.0 (#2482) [@dependabot]
• Update scipy requirement from <=1.14.0,>=1.10.1 to >=1.10.1,<=1.14.1 (#2477) [@dependabot]
• Bump pyqrack from 1.30.0 to 1.30.8 (#2476) [@dependabot]
• Bump sphinx from 7.2.6 to 8.0.2 (#2455) [@dependabot]
• Bump qibo from 0.2.9 to 0.2.10 (#2458) [@dependabot]

v0.39.0

Highlights

We've made updates to our documentation, beginning with the completion of the first section of the Pauli Twirling user guide, which offers a comprehensive introduction to this feature. Additionally, we've added a new tutorial on CDR (Clifford Data Regression) using Qrack as an efficient near-Clifford simulator. This demonstrates a workflow that harnesses the speed of Qrack in the CDR training phase, while providing users with an in-depth look at how to integrate Mitiq and Qrack effectively.

📓 Documentation

• Complete first section of Pauli Twirling user guide (#2454) [@cosenal]
• Hide primary sidebar from certain pages of the documentation (#2424) [@purva-thakre]
• CDR Tutorial with Qrack (#2451) [@bdg221]

🧑🏽‍💻 Developer Improvements

• Separate docs build workflow (#2441) [@purva-thakre]
• clean and git ignore all coverage reports (#2443) [@cosenal]
• Ignore _about.py in pytest coverage (#2379) [@purva-thakre]
• use date to make version unique for testpypi uploads (#2436) [@natestemen]

📦 Dependency Updates

• Update scipy requirement from <=1.13.1,>=1.10.1 to >=1.10.1,<=1.14.0 (#2420) [@dependabot]

(Full Changelog)

v0.38.0

Changelog

Version 0.38.0

(Full Changelog)

Highlights

🚀 As of this release, thanks to @natestemen, we are officially supporting Python 3.12 and dropping Python 3.9.

🌉 As part of UnitaryHack 2024, new contributor @NnguyenHTommy fixed a Qiskit to Cirq gate conversion error by implementing a fallback mechanism to decompose and transpile the Qiskit circuit into native Cirq gates.

🌀 Another Unitary Hacker, @EmilianoG-byte added functionality to simulate noise specifically on CNOT and CZ gates when using the Pauli Twilring technique to symmetrize errors.

🔉 As hinted in last release's spoilers, @purva-thakre has implemented the noise scaling functionality required for the Layerwise Richardson Extrapolation (LRE) technique, which allows a more fine-grained control over the amount of noise in circuits compared to the standard unitary folding method.

✨ Enhancements

• Noise Scaling for LRE (#2347) [@purva-thakre]
• Issue #2354 Fix - qiskit QFT gates error during conversion (#2404) [@NnguyenHTommy]
• Simulate noise for CNOT and CZ gates in Pauli Twirling (#2397) [@EmilianoG-byte]

🛠️ Maintenance and Upkeep Improvements

• Update readme (#2421) [@purva-thakre]
• Update README with Github discussion link (#2419) [@bdg221]
• Update contributing guide (#2382) [@purva-thakre]

🧑🏽‍💻 Dev Environment Improvements

• Used trusted publishers for testpypi publishing (#2320) [@natestemen]
• bump min python version for intersphinx map (#2425) [@natestemen]
• Support python 3.12 and drop 3.9 (#2066) [@natestemen]

📦 Dependency Updates

• Bump qibo from 0.2.8 to 0.2.9 (#2430) [@dependabot[bot]]
• Update qiskit requirement from ~=1.1.0 to ~=1.1.1 (#2414) [@dependabot[bot]]
• Bump pyscf from 2.6.0 to 2.6.2 (#2415) [@dependabot[bot]]

v0.37.0

Highlights

✨ Stacking quantum error mitigation techniques is a primary area of focus in Mitiq. In this release, @jordandsullivan introduced a Tutorial on composing Digital Dynamical Decoupling (DDD) and Zero Noise Extrapolation (ZNE).

🗒️ »Download Notebook«: Users have now the option to download tutorials in Jupyter .ipynb format directly from our documentation. We hope this will encourage experimentation with existing tutorials.

🤐 Lastly, a spoiler on what's upcoming: a Request for Comments document by @purva-thakre on adding Layerwise Noise-Scaling and Multivariate Richardson Extrapolation has been reviewed and approved. These techniques will soon make their way into Mitiq, stay tuned!

Enhancements

• Add the RFC LRE link (#2329) [@purva-thakre]
• Example stacking DDD and ZNE (#2345) [@jordandsullivan]
• Add download notebook link in docs (#2363) [@cosenal]
• Add example near-Clifford circuit simulators with links (#2367) [@bdg221]

Maintenance and upkeep improvements

• Remove qiskit-ibm-provider (#2342) [@andre-a-alves]
• Delete old version of diagram, as unnecessary (#2364) [@jordandsullivan]
• Remove duplicate readme entry in toctree (#2362) [@jordandsullivan]
• Update glossary (#2355) [@purva-thakre]
• Update QSE diagram to call module mitiq.qse, add versions to images, update references to point to v2 image (#2361) [@jordandsullivan]
• Updated docstring for generate_mirror_circuits (#2353) [@jordandsullivan]
• Pretty print supported programs enum (#2352) [@cosenal]

Dev environment improvements

• Document tip for easily generating release note (#2398) [@cosenal]
• Missing unit tests for benchmarking circuits (#2366) [@purva-thakre]

📦 Dependency updates

• Bump pyscf from 2.5.0 to 2.6.0 (#2396) [@dependabot[bot]]
• Update qiskit-aer requirement from ~=0.14.0.1 to ~=0.14.2 (#2393) [@dependabot[bot]]
• Update scipy requirement from <=1.13.0,>=1.10.1 to >=1.10.1,<=1.13.1 (#2380) [@dependabot[bot]]
• Bump qibo from 0.2.7 to 0.2.8 (#2385) [@dependabot[bot]]
• Update qiskit requirement from ~=1.0.2 to ~=1.1.0 (#2369) [@dependabot[bot]]
• Update pennylane-qiskit requirement from ~=0.35.1 to ~=0.36.0 (#2350) [@dependabot[bot]]
• Update pennylane requirement from ~=0.35.1 to ~=0.36.0 (#2349) [@dependabot[bot]]

v0.36.0

Version 0.36.0

(Full Changelog)

Highlights

Support for Qiskit 1.0: Mitiq now fully supports programs written in Qiskit 1.0, thanks to the contributions of André Alves!

Enhanced Package Requirements: We've clarified the requirements for frontend packages. Each frontend is now available as an "extra" within the Mitiq package. For instance, to use Mitiq with Qiskit, simply run:

pip install mitiq[qiskit]

and similarly for all other supported integrations.
This ensures compatibility between all dependency packages required by Mitiq for frontend integration and those in the user's environment.

Quantum Error Mitigation methods: Users can now discover the available quantum error mitigation techniques by executing:

mitiq.qem_methods()

This function provides an accessible way to understand the module naming of each technique supported by Mitiq.

Thanks to @andre-a-alves, @cosenal, @jordandsullivan, @Misty-W, @purva-thakre for the PRs in this milestone.

Enhancements

• Created method to get available QEM methods in mitiq (#2298) [@jordandsullivan]
• Define type for frontend supported programs (#2276) [@cosenal]
• Introduce requirement setup for integrations (#2303) [@cosenal]

Maintenance and upkeep improvements

• Upgrade Qiskit to 1.0.2 (#2269) [@andre-a-alves]
• Remove unused folding functions (#2289) [@jordandsullivan]
• Use Jupyter cache in gh workflow docs build (#2279) [@cosenal]
• Run linkcheck on 'release' PR workflow (#2332) [@cosenal]

Dev environment improvements

• Rename master branch to main (#2263) [@jordandsullivan]
• Pre-commit hook with style checks (#2264) [@cosenal]
• Indicate master is under development (#2258) [@natestemen]
• Remove unused devcontainer (#2302) [@cosenal]

📦 Dependency updates

• Bump myst-parser from 2.0.0 to 3.0.1 (#2333) [dependabot[bot]]
• Bump bqskit and scipy (#2262) [dependabot[bot]]
• Bump qibo from 0.2.4 to 0.2.7 (#2268) [dependabot[bot]]

🧑‍💻 Dev Dependency updates

• Update to codecov-action v4 and use token (#2271) [@cosenal]

v0.35.0

Version 0.35.0

In this milestone, we've continued our work to support Qibo by providing a new tutorial, adding related Qibo-conversion functionality to the API-doc, and added Qibo to our main list of supported frontends. We've also added the capability to use rotated randomized benchmarking circuits as part of the calibrator. These circuits provide expectation values ranging from 0 to 1 when measuring the probability that the output state is in the ground state. Having circuits with a wide range of expectation values is an important benchmarking task, and make a great test for finding the correct error mitigation technique/parameters. If you find any bugs/inconveniences in working with these updates make sure to open an issue so we are able to fix it ASAP!

This release also contains contributions from two new Mitiq contributors, and Unitary Fund team members Alessandro and Jordan! Welcome both, and looking forward to many more contributions! Well done making your first contribution so quickly 🏎️💨!

Commits

• Exit early when circuit type is not supported (#2252) [@cosenal]
• adding rotated randomized benchmarking circuits to calibrator (#2248) [@FarLab]
• add qibo + rearrange frontend order (#2249) [@purva-thakre]
• Fix examples link in README.md (#2242) [@jordandsullivan]
• add link to discussions as a way to contribute (#2234) [@natestemen]
• use python 3.11 for RTD (#2231) [@natestemen]
• fix pennylane tutorial (#2232) [@natestemen]
• Add qibo example to docs (#2220) [@francescsabater]
• replace black, flake8, and isort with ruff (#2222) [@natestemen]
• pin qibo version (#2221) [@natestemen]
• Add Qibo conversions to documentation (#2214) [@nathanshammah]
• Added warning filter to ignore warning (#2211) [@bdg221]
• remove redundant imports (#2206) [@natestemen]
• Indicate master is under development (#2205) [@natestemen]

📦 Dependency updates

• Update pennylane-qiskit requirement from ~=0.34.0 to ~=0.34.1 (#2198) [@dependabot]
• Update qiskit-ibm-provider requirement from ~=0.8.0 to ~=0.10.0 (#2196) [@dependabot]
• Update pennylane requirement from ~=0.34.0 to ~=0.35.1 (#2227) [@dependabot]
• Update pennylane-qiskit requirement from ~=0.34.1 to ~=0.35.1 (#2225) [@dependabot]
• Bump stimcirq from 1.12.1 to 1.13.0 (#2236) [@dependabot]
• Bump stim from 1.12.1 to 1.13.0 (#2237) [@dependabot]

🧑‍💻 Dev Dependency updates

• Bump all documentation dependencies (#2179) [@dependabot]
• Bump pytest-cov from 4.0.0 to 5.0.0 (#2240) [@dependabot]

v0.34.0

Changelog

Version 0.34.0

Announcing support for Qibo, a newly integrated frontend in Mitiq! 📣 Qibo is an "end-to-end open source platform for quantum simulation, self-hosted quantum hardware control, calibration and characterization".

Thank you to new contributor Francesc Sabater for excellent work integrating Qibo and Mitiq! Thanks also to new contibutor Sam Burdick for fixing our readme.

This release also includes a refactoring of part of the Mitiq shadows module, mitiq.shadows.classical_postprocessing, for speed of execution and code readability.

📓 Documentation

We've continued to enhance our (legendary!) documentation with:

1. Addition of a security policy document
2. Faster execution of the learning-based PEC tutorial in CI

Commits

• Add support for qibo circuits (#2102) [@francescsabater]
• Reduce doc build time for learning representations (#2165) [@Misty-W]
• Create SECURITY.md (#2162) [@nathanshammah]
• Fix typo in README.md (#2173) [@smburdick]
• Refactor classical postprocessing in shadows module (#2152) [@natestemen]
• Fix broken link in combine_rem_zne.md (#2156) [@Misty-W]

📦 Dependency updates

• Bump openfermion from 1.6.0 to 1.6.1 (#2182) [@dependabot]
• Update amazon-braket-sdk requirement from ~=1.68.3 to ~=1.69.0 (#2177) [@dependabot]
• Bump pyscf from 2.4.0 to 2.5.0 (#2176) [@dependabot]
• Update amazon-braket-sdk requirement from ~=1.66.0 to ~=1.68.3 (#2175) [@dependabot]
• Update qiskit-aer requirement from ~=0.13.1 to ~=0.13.2 (#2157) [@dependabot]
• Bump pytest from 7.1.3 to 8.0.0 (#2167) [@dependabot]
• Update amazon-braket-sdk requirement from ~=1.65.1 to ~=1.66.0 (#2155) [@dependabot]
• Update pennylane-qiskit requirement from ~=0.33.1 to ~=0.34.0 (#2154) [@dependabot]
• Update pennylane requirement from ~=0.33.1 to ~=0.34.0 (#2153) [@dependabot]
• Update qiskit-ibm-provider requirement from ~=0.7.3 to ~=0.8.0 (#2151) [@dependabot]

v0.33.0

Version 0.33.0

Minor update from 0.32.0 to fix a bug 🐛 in the mitiq.shadows module where an incorrect index was being used.

All changes

• Changed to use the split for b_list_shadow (#2137) [@bdg221]

v0.32.0

Version 0.32.0

Happy holidays, and happy (almost) new year!! ❄️☃️🎄🎊 This will be our last release of the year, and we'd like to thank everyone who has contributed to Mitiq over the past 12 months. We've accomplished so much in the way of error mitigation this year, and we couldn't have done it without the support and time given by the volunteers.
Thank you!

The calibrator logs have been revamped for to support result discovery and analysis. The Calibrator.run method now support two options: flat and cartesian to display the experiment results in either a linear fashion, or grid-like. Results here have been truncated for brevity.

>>> calibrator.run(log="flat")
┌──────────────────────────┬────────────────────────────────────┬────────────────────────────┐
│ benchmark                │ strategy                           │ performance                │
├──────────────────────────┼────────────────────────────────────┼────────────────────────────┤
│ Type: ghz                │ Technique: ZNE                     │ ✔                          │
│ Num qubits: 2            │ Factory: Linear                    │ Noisy error: 0.04          │
│ Circuit depth: 2         │ Scale factors: 1.0, 2.0, 3.0       │ Mitigated error: 0.02      │
│ Two qubit gate count: 1  │ Scale method: fold_gates_at_random │ Improvement factor: 2.0    │
├──────────────────────────┼────────────────────────────────────┼────────────────────────────┤
│ Type: ghz                │ Technique: ZNE                     │ ✘                          │
│ Num qubits: 2            │ Factory: Linear                    │ Noisy error: 0.04          │
│ Circuit depth: 2         │ Scale factors: 1.0, 3.0, 5.0       │ Mitigated error: 0.0658    │
│ Two qubit gate count: 1  │ Scale method: fold_global          │ Improvement factor: 0.6076 │
├──────────────────────────┼────────────────────────────────────┼────────────────────────────┤
│ Type: ghz                │ Technique: ZNE                     │ ✘                          │
│ Num qubits: 2            │ Factory: Richardson                │ Noisy error: 0.98          │
│ Circuit depth: 33        │ Scale factors: 1.0, 3.0, 5.0       │ Mitigated error: 1.03      │
│ Two qubit gate count: 14 │ Scale method: fold_global          │ Improvement factor: 0.9515 │
└──────────────────────────┴────────────────────────────────────┴────────────────────────────┘

>>> calibrator.run(log="cartesian")
┌────────────────────────────────────┬────────────────────────────┬────────────────────────────┐
│ strategy\benchmark                 │ Type: ghz                  │ Type: mirror               │
│                                    │ Num qubits: 2              │ Num qubits: 2              │
│                                    │ Circuit depth: 2           │ Circuit depth: 33          │
│                                    │ Two qubit gate count: 1    │ Two qubit gate count: 14   │
├────────────────────────────────────┼────────────────────────────┼────────────────────────────┤
│ Technique: ZNE                     │ ✘                          │ ✘                          │
│ Factory: Richardson                │ Noisy error: 0.03          │ Noisy error: 1.0           │
│ Scale factors: 1.0, 2.0, 3.0       │ Mitigated error: 0.09      │ Mitigated error: 1.03      │
│ Scale method: fold_global          │ Improvement factor: 0.3333 │ Improvement factor: 0.9709 │
├────────────────────────────────────┼────────────────────────────┼────────────────────────────┤
│ Technique: ZNE                     │ ✘                          │ ✔                          │
│ Factory: Richardson                │ Noisy error: 0.03          │ Noisy error: 1.0           │
│ Scale factors: 1.0, 3.0, 5.0       │ Mitigated error: 0.0563    │ Mitigated error: 0.97      │
│ Scale method: fold_global          │ Improvement factor: 0.5333 │ Improvement factor: 1.0309 │
├────────────────────────────────────┼────────────────────────────┼────────────────────────────┤
│ Technique: ZNE                     │ ✘                          │ ✔                          │
│ Factory: Linear                    │ Noisy error: 0.03          │ Noisy error: 1.0           │
│ Scale factors: 1.0, 3.0, 5.0       │ Mitigated error: 0.0417    │ Mitigated error: 0.9975    │
│ Scale method: fold_global          │ Improvement factor: 0.72   │ Improvement factor: 1.0025 │
└────────────────────────────────────┴────────────────────────────┴────────────────────────────┘

New benchmarking circuits: mitiq.benchmarks now contains a function generate_random_clifford_t_circuit which does what it says on the tin. Special shoutout to new UF team member Farrokh Labib (@FarLab) for this contribution.

from mitiq.benchmarks import generate_random_clifford_t_circuit

clifft = generate_random_clifford_t_circuit(
    num_qubits=2,
    num_oneq_cliffords=5,
    num_twoq_cliffords=5,
    num_t_gates=5
)
print(clifft)
# 0: ───────────@───S───T───@───H───T───X───T───T───@───@───────
#               │           │           │           │   │
# 1: ───S───T───@───────────X───────────@───S───────@───X───S───

The Executor.run method now supports a single circuit instance in addition to a list for ease of use when working with a single circuit.

- executor.run([circuit])
+ executor.run(circuit)

Faster Tests! Working on Mitiq has never been easier to develop with a faster (by 36%) test suite.

📓 Documentation

This release contains quite a few documentation improvements, including

1. New workflow images to elucidate the workflow for for using the mitiq.shadows module (available here)
2. A reorganized API-doc which should be easier to navigate
3. General clean up of the CDR user guide pages

Commits

• Clarify CDR training docs regarding the use of a markov chain monte carlo (#2130) [@natestemen]
• Update workflow images in documentation (#2034) [@purva-thakre]
• 2115 pauli twirling callibration of expectation estimation shadow needs continue (#2116) [@bdg221]
• Add randomized clifford+T benchmarking circuits. (#2118) [@FarLab]
• Frontend/Backend docs clean up (#2124) [@natestemen]
• Speed up tests using mocks (#2126) [@natestemen]
• resolve flaky mirror QV circuit test (#2127) [@natestemen + @Misty-W]
• Simplify expectation_estimation_shadow code (#2113) [@bdg221]
• Refactor calibration logs (#2074) [@kozhukalov]
• Organize API-doc (#2104) [@purva-thakre]
• add support for single circuit on exeuctor run method (#2099) [@EmilianoG-byte]

📦 Dependency updates

• Update qiskit-ibm-provider requirement from ~=0.7.2 to ~=0.7.3 (#2122) [@dependabot]
• Update cirq requirement from <1.3.0,>=1.0.0 to >=1.0.0,<1.4.0 (#2107) [@dependabot]
• Update pennylane requirement from ~=0.32.0 to ~=0.33.1 (#2091) [@dependabot]
• Update pennylane-qiskit requirement from ~=0.32.0 to ~=0.33.0 (#2081) [@dependabot]
• Bump stim/stimcirq from 1.12.0 to 1.12.1 (#2106) [@dependabot]
• Update amazon-braket-sdk requirement from ~=1.59.2 to ~=1.64.1 (#2089 + #2094 + #2114 + #2123) [@dependabot]
• Update scipy requirement from <=1.11.3,>=1.5.0 to >=1.5.0,<=1.11.4 (#2095) [@dependabot]

🧑‍💻 Dev Dependency updates

• Bump myst-nb from 0.17.1 to 1.0.0 (#2090) [@dependabot]
• Bump pandas from 2.1.2 to 2.1.3 (#2093) [@dependabot]
• Bump matplotlib from 3.8.0 to 3.8.1 (#2084) [@dependabot]
• Bump isort from 5.12.0 to 5.13.2 (#2120 + #2125) [@dependabot]
• Bump actions/setup-python from 4 to 5 (#2112) [@dependabot]