Added Dynamical QFT + Measurement by morrishuismans · Pull Request #807 · munich-quantum-toolkit/bench

morrishuismans · 2026-01-23T13:38:03Z

Description

Adds the Dynamical QFT + Measurement circuit as a dynamical benchmarking circuit.

Fixes #798

Checklist:

The pull request only contains commits that are focused and relevant to this change.
I have added appropriate tests that cover the new/changed functionality.
I have updated the documentation to reflect these changes.
I have added entries to the changelog for any noteworthy additions, changes, fixes, or removals.
I have added migration instructions to the upgrade guide (if needed).
The changes follow the project's style guidelines and introduce no new warnings.
The changes are fully tested and pass the CI checks.
I have reviewed my own code changes.

morrishuismans · 2026-01-23T13:38:52Z

@coderabbitai full review

coderabbitai · 2026-01-23T13:39:02Z

✅ Actions performed

Full review triggered.

coderabbitai · 2026-01-23T13:40:51Z

📝 Walkthrough

Summary by CodeRabbit

New Features
- Added a Dynamical Quantum Fourier Transform with Measurement benchmark circuit.
Tests
- Added a unit test validating creation and key operations of the new benchmark circuit.
Chores
- Updated Qiskit dependency requirement to 2.0.0 or newer.
Documentation
- Updated changelog and contributor list to reflect the addition.

Walkthrough

Adds a new Dynamical Quantum Fourier Transform + Measurement benchmark module, a test for a 5-qubit circuit, a changelog entry, and bumps the Qiskit dependency to >=2.0.0. The benchmark builds H gates, mid-circuit measurements, and measurement-conditioned phase rotations.

Changes

Cohort / File(s)	Summary
Dynamical QFT Benchmark Implementation `src/mqt/bench/benchmarks/dynamical_qft.py`	New module registering `create_circuit(num_qubits: int) -> QuantumCircuit` with `@register_benchmark`. Builds a circuit named `dynamical_qft`: applies H to each qubit, measures each qubit into a corresponding classical bit, and applies conditional `P(2π/2^(j+1))` rotations to subsequent qubits based on each measurement.
Benchmark Tests `tests/test_bench.py`	Added `test_dynamical_qft` that constructs a 5-qubit/5-classical-bit circuit and asserts circuit name, counts of Hadamard gates (5), measurements (5), and presence/count of conditional/if_else blocks (5).
Changelog `CHANGELOG.md`	Added an "Unreleased" entry for the Dynamical QFT + Measurement benchmark, updated PR links, and added the contributor entry.
Project Metadata `pyproject.toml`	Bumped Qiskit requirement from `qiskit[qasm3-import]>=1.3.2` to `qiskit[qasm3-import]>=2.0.0` (comment updated regarding mid-circuit measurement/support).

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

🐇 I hop through qubits with a careful chart,
H first, then I measure — a tiny art.
Bits whisper numbers, phases take flight,
Rotations twirl by classical light.
Dynamical QFT, I sing tonight! ✨

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The PR title 'Added Dynamical QFT + Measurement' directly and specifically describes the main change: introduction of a dynamical QFT circuit with measurement capability.
Description check	✅ Passed	PR description includes a clear summary, references the fixed issue `#798`, and completes all checklist items, demonstrating comprehensive coverage of required template elements.
Linked Issues check	✅ Passed	The implementation fully satisfies issue `#798` objectives: the new dynamical_qft.py module implements QFT with measurement using O(n) mid-circuit measurements and classical feed-forward instead of O(n^2) gates, with supporting tests and documentation updates.
Out of Scope Changes check	✅ Passed	The Qiskit version bump (1.3.2 to 2.0.0) is a necessary dependency update to support mid-circuit measurements required by the dynamical QFT implementation, making it in-scope.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

morrishuismans · 2026-01-23T13:46:00Z

@coderabbitai full review

coderabbitai · 2026-01-23T13:46:06Z

✅ Actions performed

Full review triggered.

coderabbitai

Actionable comments posted: 3

🤖 Fix all issues with AI agents

In `@src/mqt/bench/benchmarks/dynamical_qft.py`:
- Line 21: The description string passed to the register_benchmark decorator for
"dynamical_qft" contains a leading space; update the decorator usage (the
`@register_benchmark` call for "dynamical_qft") to remove the leading space so the
description reads "Dynamic Quantum Fourier Transformation and Measurement (DQFT
+ M)" instead of " Dynamic...".

In `@tests/test_bench.py`:
- Around line 227-234: Add a circuit name assertion to test_dynamical_qft to
match other tests: after creating qc via create_circuit("dynamical_qft", 5)
assert that qc.name (or the circuit's name property used elsewhere) equals
"dynamical_qft" to ensure consistency with tests like test_bv and
test_graphstate_seed; update the test_dynamical_qft function to include this
assertion immediately after the circuit is created.
- Around line 236-245: The test_dynamical_qft_gates currently only checks
Hadamard and measurement counts; add an assertion to verify the conditional
phase rotations by counting entries in qc.data where inst.operation.name is "p"
or the phase gate identifier used by create_circuit for dynamical_qft, and
assert that for num_qubits=5 the phase/controlled-phase gate count equals 10
(4+3+2+1+0). Locate the test function test_dynamical_qft_gates and use qc.data
iteration (same pattern as hadamard_count) to compute phase_count and add the
assert phase_count == 10 so the dynamical QFT's conditional rotations are
covered.

src/mqt/bench/benchmarks/dynamical_qft.py

tests/test_bench.py

codecov · 2026-01-23T16:30:52Z

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

morrishuismans · 2026-01-23T17:11:41Z

@coderabbitai full review

coderabbitai · 2026-01-23T17:11:51Z

✅ Actions performed

Full review triggered.

coderabbitai

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@pyproject.toml`:
- Around line 22-24: The pyproject currently pins "qiskit[qasm3-import]>=2.0.0"
while the comment references fixes present in Qiskit 1.3.2; either lower the
minimum to the appropriate older version (e.g., "qiskit[qasm3-import]>=1.3.2" or
the exact earliest release containing the OpenQASM3 export fixes) or add a clear
justification next to the dependency line explaining why 2.0.0 is required
(mentioning the specific features like BoxOp or control-flow changes if relied
upon); update the dependency string "qiskit[qasm3-import]>=2.0.0" or the
adjacent comment accordingly.

pyproject.toml

morrishuismans · 2026-01-28T17:23:00Z

@flowerthrower @burgholzer Dear Lukas and Patrick, could you please review my code? Thanks in advance!

Signed-off-by: morrishuismans <97433985+morrishuismans@users.noreply.github.com>

burgholzer

Thanks for your work on this!
This looks like great progress already.
You can find some suggestions/feedback in the comments below.
I hope none of them should really be hard to address.

You will also need to merge in the latest changes from main so that the conflicts are resolved.

burgholzer · 2026-02-04T08:17:13Z

src/mqt/bench/benchmarks/dynamical_qft.py

+
+# QFT & M v.1


Suggested change

# QFT & M v.1

The docstring here feels a bit arbitrary

burgholzer · 2026-02-04T08:19:12Z

src/mqt/bench/benchmarks/dynamical_qft.py

+
+# QFT & M v.1
+
+"""Dynamical QFT + M benchmark definition."""


I would have a slight tendency to consistently refer to this as Dynamic QFT (instead of "dynamical" or "QFT+M"). This applies to all occurrences where the name is used.

burgholzer · 2026-02-04T08:20:51Z

src/mqt/bench/benchmarks/dynamical_qft.py

+
+@register_benchmark("dynamical_qft", description="Dynamic Quantum Fourier Transformation and Measurement (DQFT + M)")
+def create_circuit(num_qubits: int) -> QuantumCircuit:
+    """Returns a quantum circuit implementing the Dynamic Quantum Fourier Transform and Measurement algorithm.


This docstring could be made a little more detailed and explain a little bit more about the algorithm itself. From this docstring alone, no one would have an idea of what the algorithm is doing. There is also no reference to some kind of resource this is based on.

burgholzer · 2026-02-04T08:23:50Z

src/mqt/bench/benchmarks/dynamical_qft.py

+    Arguments:
+        num_qubits: number of qubits of the returned quantum circuit
+
+    Returns:
+        QuantumCircuit: a quantum circuit implementing the Dynamic Quantum Fourier Transform and Measurement algorithm
+    """
+    q = QuantumRegister(num_qubits, "q")
+    cl = ClassicalRegister(num_qubits, "cl")
+    qc = QuantumCircuit(q, cl, name="dynamical_qft")
+    for i in range(num_qubits):
+        qc.h(q[i])
+        qc.measure(q[i], cl[i])
+        with qc.if_test((cl[i], 1)):
+            for j in range(1, num_qubits - i):
+                qc.p(2 * np.pi / 2 ** (j + 1), q[j + i])
+    return qc


Initially I was thinking this isn't really the way I imagined this to be written, because I would have imagined this to be a single-qubit circuit that repeatedly resets the qubit and applies phase corrections to it based on previous measurements (similar to iQPE from #806).
However, after thinking about it a little more, this kind of does make sense to portray as an n qubit circuit. This also allows using this circuit as a benchmark for circuit optimization techniques that would, e.g., introduce qubit reuse to shrink the circuit.

Micro optimization: The phase computation can be simplified a bit here

qc.p(2 * np.pi / 2 ** (j + 1), q[j + I])

is equivalent to

qc.p(np.pi / (2 ** j), q[j + I])

Slight naming tweak: The canonical name for classical registers in Qiskit is c (not cl) might be worth adapting that.

Similar to #806, this will need an upper bound on the number of qubits since the shift operation only makes sense up to a certain number of qubits and is not well defined afterwards.

Added Dynamical QFT + Measurement

e34849e

github-project-automation bot added this to MQT Applications Jan 23, 2026

github-project-automation bot moved this to In Progress in MQT Applications Jan 23, 2026

Merge branch 'main' into dynamic_QFT

33eda71

🎨 pre-commit fixes

febd72c

coderabbitai bot requested changes Jan 23, 2026

View reviewed changes

src/mqt/bench/benchmarks/dynamical_qft.py Outdated Show resolved Hide resolved

tests/test_bench.py Show resolved Hide resolved

tests/test_bench.py Outdated Show resolved Hide resolved

morrishuismans added 2 commits January 23, 2026 17:05

Qiskit 2.0.0

2984c7c

Dynamical QFT + Measurement with coderabbit fixes

ec58a0f

coderabbitai bot approved these changes Jan 23, 2026

View reviewed changes

🎨 pre-commit fixes

a7c9463

CHANGELOG.md updated

30f31dd

coderabbitai bot requested changes Jan 23, 2026

View reviewed changes

pyproject.toml Show resolved Hide resolved

morrishuismans marked this pull request as ready for review January 27, 2026 08:35

morrishuismans marked this pull request as draft January 27, 2026 08:36

morrishuismans marked this pull request as ready for review January 28, 2026 17:19

coderabbit fixes 2

a937af4

morrishuismans marked this pull request as draft February 2, 2026 14:25

coderabbitai bot approved these changes Feb 2, 2026

View reviewed changes

Merge branch 'main' into dynamic_QFT

92f0517

Signed-off-by: morrishuismans <97433985+morrishuismans@users.noreply.github.com>

morrishuismans marked this pull request as ready for review February 2, 2026 16:10

burgholzer self-requested a review February 3, 2026 20:36

burgholzer added python Pull requests that update Python code feature New feature or request labels Feb 4, 2026

burgholzer requested changes Feb 4, 2026

View reviewed changes

Uh oh!

Conversation

morrishuismans commented Jan 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist:

Uh oh!

morrishuismans commented Jan 23, 2026

Uh oh!

coderabbitai bot commented Jan 23, 2026

Uh oh!

coderabbitai bot commented Jan 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Walkthrough

Changes

Estimated code review effort

Poem

Uh oh!

morrishuismans commented Jan 23, 2026

Uh oh!

coderabbitai bot commented Jan 23, 2026

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

codecov bot commented Jan 23, 2026

Codecov Report

Uh oh!

morrishuismans commented Jan 23, 2026

Uh oh!

coderabbitai bot commented Jan 23, 2026

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

morrishuismans commented Jan 28, 2026

Uh oh!

burgholzer left a comment

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

burgholzer Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

morrishuismans commented Jan 23, 2026 •

edited

Loading

coderabbitai bot commented Jan 23, 2026 •

edited

Loading