Circuit cutting: add mid-circuit measurement integration tests #2234

trbromley · 2022-02-25T22:25:02Z

Context:

Adds tests for the cut_circuit transform that ensures circuits resulting in fragments that contain mid-circuit measurements are supported.

github-actions · 2022-02-25T22:25:21Z

Hello. You may have forgotten to update the changelog!
Please edit doc/releases/changelog-dev.md with:

A one-to-two sentence description of the change. You may include a small working example for new features.
A link back to this PR.
Your name (or GitHub username) in the contributors section.

trbromley · 2022-02-25T22:30:31Z

pennylane/transforms/qcut.py

+    copy_ops = [copy.copy(op) for _, op in ordered_ops if not isinstance(op, MeasurementProcess)]
+    copy_meas = [copy.copy(op) for _, op in ordered_ops if isinstance(op, MeasurementProcess)]


Previously we were treating the ops (gates) and measurements together. However, when isinstance(op, MeasurementProcess), then setting op._wires = ... does not work when the MeasurementProcess is an expectation value of an observable, because the measurement process uses the contained observable to determine op.wires.

In this updated version, we treat the MeasurementProcesses more carefully after applying the gates.

Nice!

I'm curious what where the cases that revealed this? Mid circuit measurements and tensor products of obs in measurements?

Right, previously if your observable (or part of it) was on wire i, but wire i had mid-circuit measurements and was relabelled to wire i', then the observable wasn't getting its wire updated from wire i to i'.

pennylane/transforms/qcut.py

trbromley · 2022-02-25T22:33:46Z

pennylane/transforms/qcut.py

@@ -424,7 +441,7 @@ def _get_measurements(

    obs = measurement.obs

-    return [expval(obs @ g) for g in group]
+    return [expval(copy.deepcopy(obs) @ g) for g in group]


I am very confused by this. the tensor product works in-place 🤔 Perhaps it is a queuing-related quirk.

obs = qml.PauliZ(0) @ qml.PauliZ(1) obs2 = obs @ qml.PauliZ(2) assert obs == obs2 print(obs)

I see, so the original obs is being overwritten in when we take a tensor product of it with another tensor. This is weird!

But deepcopy seems to solve for our needs, or is there potential for more problems with this?

Also works with copy.copy so I changed to that. I think copy.copy is a sufficient workaround for now, but long run it'd be better for observables not to be altered when taking the tensor product. Issue posted: #2235

codecov · 2022-02-25T22:50:45Z

Codecov Report

Merging #2234 (bb99ee3) into master (7ad040b) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #2234   +/-   ##
=======================================
  Coverage   99.26%   99.27%           
=======================================
  Files         231      231           
  Lines       18349    18359   +10     
=======================================
+ Hits        18215    18225   +10     
  Misses        134      134

Impacted Files	Coverage Δ
pennylane/transforms/qcut.py	`100.00% <100.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7ad040b...bb99ee3. Read the comment docs.

trbromley · 2022-02-25T22:51:00Z

[sc-14724]

anthayes92

Great spots here! The pipeline is becoming robust 🦾

Would be interested to find out whats going on with the tensor product quirk you've found 🤔

anthayes92 · 2022-02-26T16:47:51Z

tests/transforms/test_qcut.py

+            qml.CNOT(wires=[0, 1])
+            return qml.expval(qml.PauliZ(0) @ qml.PauliZ(1))
+
+        spy = mocker.spy(qcut, "qcut_processing_fn")


perhaps it would be more appropriate to spy on _get_measurements in this test

Happy to change, but why do you think _get_measurements is more appropriate?

anthayes92 · 2022-02-26T16:55:39Z

pennylane/transforms/qcut.py

@@ -424,7 +441,7 @@ def _get_measurements(

    obs = measurement.obs

-    return [expval(obs @ g) for g in group]
+    return [expval(copy.deepcopy(obs) @ g) for g in group]


I see, so the original obs is being overwritten in when we take a tensor product of it with another tensor. This is weird!

But deepcopy seems to solve for our needs, or is there potential for more problems with this?

pennylane/transforms/qcut.py

anthayes92 · 2022-02-26T17:00:09Z

pennylane/transforms/qcut.py

+    copy_ops = [copy.copy(op) for _, op in ordered_ops if not isinstance(op, MeasurementProcess)]
+    copy_meas = [copy.copy(op) for _, op in ordered_ops if isinstance(op, MeasurementProcess)]


Nice!

I'm curious what where the cases that revealed this? Mid circuit measurements and tensor products of obs in measurements?

… qcut_integ_tests

trbromley

Thanks @anthayes92!

trbromley · 2022-02-28T13:46:32Z

pennylane/transforms/qcut.py

+    copy_ops = [copy.copy(op) for _, op in ordered_ops if not isinstance(op, MeasurementProcess)]
+    copy_meas = [copy.copy(op) for _, op in ordered_ops if isinstance(op, MeasurementProcess)]


Right, previously if your observable (or part of it) was on wire i, but wire i had mid-circuit measurements and was relabelled to wire i', then the observable wasn't getting its wire updated from wire i to i'.

trbromley · 2022-02-28T13:49:07Z

pennylane/transforms/qcut.py

@@ -424,7 +441,7 @@ def _get_measurements(

    obs = measurement.obs

-    return [expval(obs @ g) for g in group]
+    return [expval(copy.deepcopy(obs) @ g) for g in group]


Also works with copy.copy so I changed to that. I think copy.copy is a sufficient workaround for now, but long run it'd be better for observables not to be altered when taking the tensor product. Issue posted: #2235

trbromley · 2022-02-28T13:51:55Z

tests/transforms/test_qcut.py

+            qml.CNOT(wires=[0, 1])
+            return qml.expval(qml.PauliZ(0) @ qml.PauliZ(1))
+
+        spy = mocker.spy(qcut, "qcut_processing_fn")


Happy to change, but why do you think _get_measurements is more appropriate?

trbromley added 12 commits February 25, 2022 12:46

Add test

73acf20

Add to changelog

11c9091

Add tests in different interfaces

447af24

Work on TF Jit

a88da66

Remove JIT tests

b4314e4

Fix bug in changing wires of MeasurementProcesses

f3e8a56

Fix graph_to_tape

5b69cb5

Add supporting tests

f0f3839

Parametrize over use_opt_einsum

7192f3d

Merge branch 'qcut_integ_tests' into qcut_integ_tests_2

0cc78d7

Copy observable to prevent in-place modification

4ac5f7b

Add test

6f68b95

trbromley added 2 commits February 25, 2022 17:25

Remove added line

89dd999

Add to changelog

01e507c

trbromley commented Feb 25, 2022

View reviewed changes

trbromley requested review from anthayes92 and angusjlowe February 25, 2022 22:36

anthayes92 approved these changes Feb 26, 2022

View reviewed changes

anthayes92 and others added 9 commits February 26, 2022 12:06

Merge branch 'master' into qcut_integ_tests

b4a9d66

Remove mention of jit

23e7c63

Merge branch 'qcut_integ_tests' into qcut_integ_tests_2

c7b8532

Remove isinstance check

9bf43a6

Merge branch 'qcut_integ_tests' into qcut_integ_tests_2

57c247d

Merge branch 'qcut_integ_tests' of github.com:XanaduAI/pennylane into…

3d39cb9

… qcut_integ_tests

Merge branch 'master' into qcut_integ_tests

fe71c95

Merge branch 'qcut_integ_tests' into qcut_integ_tests_2

9554fe8

Switch from deepcopy to copy

55da371

trbromley commented Feb 28, 2022

View reviewed changes

Base automatically changed from qcut_integ_tests to master February 28, 2022 14:02

Merge branch 'master' into qcut_integ_tests_2

bb99ee3

trbromley merged commit 57237b2 into master Feb 28, 2022

trbromley deleted the qcut_integ_tests_2 branch February 28, 2022 20:06

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Circuit cutting: add mid-circuit measurement integration tests #2234

Circuit cutting: add mid-circuit measurement integration tests #2234

trbromley commented Feb 25, 2022

github-actions bot commented Feb 25, 2022

trbromley Feb 25, 2022

anthayes92 Feb 26, 2022

trbromley Feb 28, 2022

trbromley Feb 25, 2022

anthayes92 Feb 26, 2022 •

edited

Loading

trbromley Feb 28, 2022

codecov bot commented Feb 25, 2022 •

edited

Loading

trbromley commented Feb 25, 2022

anthayes92 left a comment

anthayes92 Feb 26, 2022

trbromley Feb 28, 2022

anthayes92 Feb 26, 2022 •

edited

Loading

anthayes92 Feb 26, 2022

trbromley left a comment

trbromley Feb 28, 2022

trbromley Feb 28, 2022

trbromley Feb 28, 2022

		copy_ops = [copy.copy(op) for _, op in ordered_ops if not isinstance(op, MeasurementProcess)]
		copy_meas = [copy.copy(op) for _, op in ordered_ops if isinstance(op, MeasurementProcess)]

Circuit cutting: add mid-circuit measurement integration tests #2234

Circuit cutting: add mid-circuit measurement integration tests #2234

Conversation

trbromley commented Feb 25, 2022

github-actions bot commented Feb 25, 2022

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

anthayes92 Feb 26, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codecov bot commented Feb 25, 2022 • edited Loading

Codecov Report

trbromley commented Feb 25, 2022

anthayes92 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

anthayes92 Feb 26, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

trbromley left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

anthayes92 Feb 26, 2022 •

edited

Loading

codecov bot commented Feb 25, 2022 •

edited

Loading

anthayes92 Feb 26, 2022 •

edited

Loading