Implement cutting of general 2-qubit unitaries #302

garrison · 2023-07-02T02:23:59Z

Builds on #294. Closes #186. ~~I'm marking as "on hold" at least until we merge the other explicit gate PRs.~~

Remaining action items

Release note
Rename supported_gates() -> explicitly_supported_gates()
update docstring
More comments/explanation
Update references in readme and docs

[ci skip]

and can be singletons, one day!

Builds on #294. Closes #186.

circuit_knitting/cutting/qpd/qpd.py

caleb-johnson

So happy you figured this all out. I'm approving bc tests look good, but I'll be interested to hear how you connected the dots between the literature and the WeylDecomposer class in Qiskit. Thanks!

caleb-johnson · 2023-07-03T15:51:41Z

Maybe not this PR, but should we provide a lookup table for Qiskit gates --> Sampling overhead, now that we suddenly support a lot of gates. Also, we now have a grab bag of optimal and non-optimal decomps in our supported gate set. How do we make this clear? Maybe two sections in the lookup table?

PR still in draft status

garrison · 2023-07-05T19:01:23Z

Maybe not this PR, but should we provide a lookup table for Qiskit gates --> Sampling overhead

I agree -- it is worth having a section about this in the documentation. I could imagine this going in either the "explanation" section or the API references. It feels like a "reference"-type documentation to me, even though it is not strictly about the API.

Also, we now have a grab bag of optimal and non-optimal decomps in our supported gate set. How do we make this clear? Maybe two sections in the lookup table?

The authors of https://arxiv.org/abs/2006.11174v2 left open the possibility that their decomposition might not actually be optimal. (They discuss this in Sec. 2.4.) However, corollary 4.4 of https://arxiv.org/abs/2205.00016v2 actually shows that each of the gates that we explicitly support has an optimal decomposition. So, we are really left with two classes: decompositions that we can prove are optimal, and decompositions that may be optimal. Pretty much everything we explicitly mention in our table is going to fit in the first category; I can't think of a single "standard" gate that actually fits in the second. EDIT: actually, XXPlusYYGate and XXMinusYYGate might fit in the second category. I've so far been unable to work out a closed form expression for their overhead. (I haven't tried very hard yet, but I did try briefly.)

garrison · 2023-07-05T20:59:02Z

We might as well put this on the 0.3 milestone, right? Deciding which release will influence whether I modify existing release notes or add new ones.

caleb-johnson · 2023-07-05T21:37:19Z

We might as well put this on the 0.3 milestone, right? Deciding which release will influence whether I modify existing release notes or add new ones.

Yes I thnk so

coveralls · 2023-07-19T21:48:18Z

Pull Request Test Coverage Report for Build 5663627358

27 of 27 (100.0%) changed or added relevant lines in 1 file are covered.
2 unchanged lines in 1 file lost coverage.
Overall coverage increased (+0.08%) to 90.794%

Files with Coverage Reduction	New Missed Lines	%
circuit_knitting/utils/orbital_reduction.py	2	60.0%

Totals
Change from base Build 5590764933:	0.08%
Covered Lines:	2584
Relevant Lines:	2846

💛 - Coveralls

caleb-johnson · 2023-07-20T20:22:41Z

circuit_knitting/cutting/__init__.py

@@ -57,7 +57,7 @@
    qpd.generate_qpd_weights
    qpd.generate_qpd_samples
    qpd.decompose_qpd_instructions
-    qpd.supported_gates
+    qpd.explicitly_supported_gates


I wanted to suggest optimally_supported_gates, but I can't convince myself I like it more. I think they are all optimal, at least?

Yeah, they are all optimal -- even the ones for which we do not have explicit support.

Another option could be to get rid of this function -- or at least to make it private. The reason it was introduced in #277 was to support #278, but now that all two-qubit gates can be cut, it's no longer necessary. It might be nice to have something like this if we one day support some (but not all) 3-qubit gates (#258), which could be a case for keeping it but prefixing it with an underscore. This way, we'd at least have a use case in mind (again) before committing to support it.

By the way, all essential tests (i.e., all but a few specific ones) work if we take this PR and remove all the explicit gate support, i.e., if we apply the further patch:

diff --git a/circuit_knitting/cutting/qpd/qpd.py b/circuit_knitting/cutting/qpd/qpd.py index 1e5be49..c6974a8 100644 --- a/circuit_knitting/cutting/qpd/qpd.py +++ b/circuit_knitting/cutting/qpd/qpd.py @@ -573,13 +573,6 @@ def qpdbasis_from_gate(gate: Gate) -> QPDBasis: ValueError: Cannot decompose gate with unbound parameters. ValueError: ``to_matrix`` conversion of two-qubit gate failed. """ - try: - f = _qpdbasis_from_gate_funcs[gate.name] - except KeyError: - pass - else: - return f(gate) - if isinstance(gate, Gate) and gate.num_qubits == 2: try: mat = gate.to_matrix()

so explicit support doesn't mean a ton anyway.

With #174, there will be one additional explicitly supported instruction: the Move instruction, which is not a Gate because it is not unitary. So it might be nice to have a function like the current one, somewhere. I'm leaning toward making the whole function private for now and removing it from the release notes until we better understand how we expect it to be used.

The more I thought about it, the more I felt like the most reasonable thing to do is to remove this function from the public API for now, hence my change in 8e95c4a.

It's not clear that this function remains useful now that we support essentially all 2-qubit gates. If we find a use for it in the future, we can re-introduce it (or something like it) as a public interface.

caleb-johnson · 2023-07-25T22:23:46Z

circuit_knitting/cutting/qpd/__init__.py

@@ -18,7 +18,6 @@
    decompose_qpd_instructions,
    WeightType,
    qpdbasis_from_gate,
-    supported_gates,


Is there a 2-qubit gate in Qiskit that would break our cutting workflow if we tried to cut it?

I'm wondering if we can replace
if g in supported_gates():

with
if g.num_qubits == 2:

I now see your docstring update below. They have to implement to_matrix

caleb-johnson

Nice work and great documentation, thanks! LGTM 👍

caleb-johnson · 2023-07-25T22:25:14Z

circuit_knitting/cutting/qpd/__init__.py

@@ -18,7 +18,6 @@
    decompose_qpd_instructions,
    WeightType,
    qpdbasis_from_gate,
-    supported_gates,


I now see your docstring update below. They have to implement to_matrix

caleb-johnson · 2023-07-25T22:29:16Z

circuit_knitting/cutting/qpd/qpd.py

+    theta: np.typing.NDArray[np.float64] | Sequence[float], /
+) -> np.typing.NDArray[np.complex128]:
+    r"""
+    Exponentiate the non-local portion of a KAK decomposition.


Great writeup here, thanks

* Add support for `SwapGate` * Reorder terms [ci skip] * Add missing terms * DRY the coefficients * Fix coverage * Add support for `iSwapGate` * Fix black * Add to release note * Fix type hint * Gates without parameters are nicer to work with and can be singletons, one day! * Remove a line * Add comments describing channels * `_copy_unique_sublists` * Add `DCXGate` * Tweak * Implement cutting of general 2-qubit unitaries Builds on #294. Closes #186. * Add tests of additional gates * Fix type annotation * Add explanatory comments * `supported_gates()` -> `explicitly_supported_gates()` * Add to references * Improved error message and test about `to_matrix` conversion failing * Add xref to `QPDBasis` in docstrings * Add `qpdbasis_from_gate` to Sphinx build * Make `explicitly_supported_gates` private and remove its release note It's not clear that this function remains useful now that we support essentially all 2-qubit gates. If we find a use for it in the future, we can re-introduce it (or something like it) as a public interface. * Fix intersphinx link * Release note * Update qpd.py: remove extraneous `from None`

* Don't batch unnecessarily * mypy * black * Missing varname * fix bug * Clean up code * Improve comments * release note * release note * Update batch-by-sampler-c4ae836df9997b1d.yaml * Update batch-by-sampler-c4ae836df9997b1d.yaml * Update batch-by-sampler-c4ae836df9997b1d.yaml * Bump Python version in Dockerfile to 3.11 (#331) Now that CKT supports Python 3.11, we might as well use the latest version in the Dockerfile. * Fix formatting of example in `reduce_bitstrings` docstring (#332) * Implement cutting of general 2-qubit unitaries (#302) * Add support for `SwapGate` * Reorder terms [ci skip] * Add missing terms * DRY the coefficients * Fix coverage * Add support for `iSwapGate` * Fix black * Add to release note * Fix type hint * Gates without parameters are nicer to work with and can be singletons, one day! * Remove a line * Add comments describing channels * `_copy_unique_sublists` * Add `DCXGate` * Tweak * Implement cutting of general 2-qubit unitaries Builds on #294. Closes #186. * Add tests of additional gates * Fix type annotation * Add explanatory comments * `supported_gates()` -> `explicitly_supported_gates()` * Add to references * Improved error message and test about `to_matrix` conversion failing * Add xref to `QPDBasis` in docstrings * Add `qpdbasis_from_gate` to Sphinx build * Make `explicitly_supported_gates` private and remove its release note It's not clear that this function remains useful now that we support essentially all 2-qubit gates. If we find a use for it in the future, we can re-introduce it (or something like it) as a public interface. * Fix intersphinx link * Release note * Update qpd.py: remove extraneous `from None` * Improve the instructions regarding pandoc (#336) I've also tried to make the developer documentation easier to discover * Make the repository link more obvious from the Sphinx build (#338) * Make the repository link more obvious from the Sphinx build * Enable "edit" link in the header * Add comment * Add README badge linking to stable documentation (#339) * Add docs badge to link to stable docs * Add ruff badge * Remove ruff * Update README.md (#340) Making directions on opening the docs a little more succinct. * Change var name * Add SECURITY.md (#337) * Implement wire cutting as a two-qubit instruction (#174) * Implement wire cutting as a two-qubit instruction * Update type annotation * s/gate/instruction/ * Add overhead test for `Move` instruction * Add wire cutting tutorial * Add `Move` to Sphinx build * Doc updates suggested by Caleb * Add release note and link to new tutorial * Clarify wording following #174 (comment) * Improvements to `Move` docstring * Use svg as the plot format This avoids pixelation on high-dpi displays * Remove unnecessary uses of `CircuitInstruction` https://github.com/Qiskit-Extensions/circuit-knitting-toolbox/pull/174/files#r1278067109 * The notebook tests should ignore any files that crop up in `docs/_build` `matplotlib.sphinxext.plot_directive` likes to leave python files there --------- Co-authored-by: Jim Garrison <garrison@ibm.com> Co-authored-by: Ibrahim Shehzad <75153717+IbrahimShehzad@users.noreply.github.com>

garrison and others added 17 commits June 29, 2023 23:02

Add support for SwapGate

b9d7aed

Reorder terms

779525b

[ci skip]

Add missing terms

56a4419

Merge branch 'main' into swap-gate

a65397f

DRY the coefficients

3edcc96

Fix coverage

b114c0d

Add support for iSwapGate

489b68d

Fix black

3515e0e

Add to release note

1c95367

Fix type hint

040f219

Gates without parameters are nicer to work with

74e5a9d

and can be singletons, one day!

Remove a line

f8aa065

Add comments describing channels

e0a5f27

_copy_unique_sublists

d790f1d

Add DCXGate

a9ec8b6

Tweak

69a1e69

Implement cutting of general 2-qubit unitaries

727d159

Builds on #294. Closes #186.

garrison added enhancement New feature or request cutting QPD-based circuit cutting code labels Jul 2, 2023

garrison added this to the 0.4.0 milestone Jul 2, 2023

garrison added the on hold Let's wait for something before merging label Jul 2, 2023

garrison added 2 commits July 2, 2023 22:07

Add tests of additional gates

59b2691

Fix type annotation

9a7121a

caleb-johnson reviewed Jul 3, 2023

View reviewed changes

circuit_knitting/cutting/qpd/qpd.py Show resolved Hide resolved

caleb-johnson previously approved these changes Jul 3, 2023

View reviewed changes

Base automatically changed from swap-gate to main July 5, 2023 20:22

Merge branch 'main' into general-unitaries

4232864

Add explanatory comments

05755e3

garrison modified the milestones: 0.4.0, 0.3.0 Jul 5, 2023

garrison removed the on hold Let's wait for something before merging label Jul 6, 2023

garrison mentioned this pull request Jul 6, 2023

Test cutting of CSXGate().inverse() #309

Closed

Merge branch 'main' into general-unitaries

39b865e

garrison added 2 commits July 20, 2023 15:21

supported_gates() -> explicitly_supported_gates()

0823787

Add to references

236fad2

caleb-johnson reviewed Jul 20, 2023

View reviewed changes

garrison added 5 commits July 24, 2023 13:54

Improved error message and test about to_matrix conversion failing

874d1a3

Add xref to QPDBasis in docstrings

bee23a8

Add qpdbasis_from_gate to Sphinx build

f43d474

Make explicitly_supported_gates private and remove its release note

8e95c4a

It's not clear that this function remains useful now that we support essentially all 2-qubit gates. If we find a use for it in the future, we can re-introduce it (or something like it) as a public interface.

Fix intersphinx link

3bfccfc

garrison linked an issue Jul 24, 2023 that may be closed by this pull request

Support a wider variety of Qiskit gates for automatic, optimal decomposition #288

Closed

11 tasks

Release note

51a9674

garrison marked this pull request as ready for review July 24, 2023 22:16

garrison mentioned this pull request Jul 24, 2023

Support a wider variety of Qiskit gates for automatic, optimal decomposition #288

Closed

11 tasks

caleb-johnson reviewed Jul 25, 2023

View reviewed changes

caleb-johnson previously approved these changes Jul 25, 2023

View reviewed changes

Update qpd.py: remove extraneous from None

91ac6b7

garrison dismissed caleb-johnson’s stale review via 91ac6b7 July 26, 2023 01:35

garrison merged commit d1db839 into main Jul 26, 2023
9 checks passed

garrison deleted the general-unitaries branch July 26, 2023 01:43

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implement cutting of general 2-qubit unitaries #302

Implement cutting of general 2-qubit unitaries #302

garrison commented Jul 2, 2023 •

edited

Loading

caleb-johnson left a comment

caleb-johnson commented Jul 3, 2023 •

edited

Loading

garrison commented Jul 5, 2023 •

edited

Loading

garrison commented Jul 5, 2023

caleb-johnson commented Jul 5, 2023

coveralls commented Jul 19, 2023 •

edited

Loading

caleb-johnson Jul 20, 2023

garrison Jul 21, 2023 •

edited

Loading

garrison Jul 24, 2023

caleb-johnson Jul 25, 2023

caleb-johnson Jul 25, 2023

caleb-johnson left a comment

caleb-johnson Jul 25, 2023

caleb-johnson Jul 25, 2023

Implement cutting of general 2-qubit unitaries #302

Implement cutting of general 2-qubit unitaries #302

Conversation

garrison commented Jul 2, 2023 • edited Loading

Remaining action items

caleb-johnson left a comment

Choose a reason for hiding this comment

caleb-johnson commented Jul 3, 2023 • edited Loading

garrison commented Jul 5, 2023 • edited Loading

garrison commented Jul 5, 2023

caleb-johnson commented Jul 5, 2023

coveralls commented Jul 19, 2023 • edited Loading

Pull Request Test Coverage Report for Build 5663627358

💛 - Coveralls

caleb-johnson Jul 20, 2023

Choose a reason for hiding this comment

garrison Jul 21, 2023 • edited Loading

Choose a reason for hiding this comment

garrison Jul 24, 2023

Choose a reason for hiding this comment

caleb-johnson Jul 25, 2023

Choose a reason for hiding this comment

caleb-johnson Jul 25, 2023

Choose a reason for hiding this comment

caleb-johnson left a comment

Choose a reason for hiding this comment

caleb-johnson Jul 25, 2023

Choose a reason for hiding this comment

caleb-johnson Jul 25, 2023

Choose a reason for hiding this comment

garrison commented Jul 2, 2023 •

edited

Loading

caleb-johnson commented Jul 3, 2023 •

edited

Loading

garrison commented Jul 5, 2023 •

edited

Loading

coveralls commented Jul 19, 2023 •

edited

Loading

garrison Jul 21, 2023 •

edited

Loading