Fixed Qdrift test

[Jaybsoni]

Context:
Integration tests for the Qdrift template were failing. These tests were comparing the state produced after applying QDrift with the state produced after applying its decomposition.

The source of the discrepancy was a bug in the tests themselves. Specifically, the expected state was produced by multiplying the gates in the expected decomposition of QDrift in the opposite order. The order should be reversed from that of the decomposition.

This should have caused all of the tests to fail initially, however the hamiltonians used while testing Qdrift had accidentally been chosen such that most of their terms commuted with each other. Due to this, the order of the decomposition didn't matter except for the few edge cases.

Description of the Change:

• Updated the test hamiltonians so that they don't all commute.
• Reversed the order of the expected_decomp list

[github-actions]

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.

[Jaybsoni]

[sc-91298]

[Jaybsoni]

Since the bug was in the test itself, I think we don't need a changelog entry for this correct?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.68%. Comparing base (05b469d) to head (d3e8ce5).
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #7621   +/-   ##
=======================================
  Coverage   99.68%   99.68%           
=======================================
  Files         530      530           
  Lines       52140    52140           
=======================================
  Hits        51977    51977           
  Misses        163      163           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[JerryChen97]

Thank you so much for fixing this! @Jaybsoni
On the changelog, I guess we can put a note in the internal part. Usually if it's a very minor fix to test suite or something like typo correction we don't put into changelog atl, but for things like fixing known issues within test suites we still leave a note there.
Also, mind creating a new test to specifically test it with a Hamiltonian that from the very beginning won't allow similar coincidental order reversal atl?

[Jaybsoni]

@JerryChen97

Also, mind creating a new test to specifically test it with a Hamiltonian that from the very beginning won't allow similar coincidental order reversal atl?

I changed the hamiltonian on line 35 such that its terms no longer commute. Furthermore, the hamiltonian on line 31 also has terms which don't commute. I think this should be sufficient. If we want, we can increase the number of samples (n), this would increase the likelihood of sampling terms which don't commute and would have cause the test to fail sooner. However I don't think it's required anymore.

[JerryChen97]

LGTM!

[SimoneGasperini]

LGTM!

I don't have anything else to say. I was about to suggest just adding a comment, making sure to use non-commuting terms in the test Hamiltonian, but I see that it is already there.

Thank you!