Gate set tomography

[mho291]

One qubit calibration (empty circuit tomography)
Two qubit calibration (empty circuit tomography)
One qubit operator(s) gate set tomography
Two qubits operator(s) gate set tomography

Checklist:

• Reviewers confirm new code works as expected.
• Tests are passing.
• Coverage does not decrease.
• Documentation is updated.
• Testing with QiboLab backend.

[BrunoLiegiBastonLiegi]

Thanks @mho291 for implementing this. I left some comments below, I'm not finished yet, but in general I have the feeling that many code repetitions can be avoided. I would try to write some general functions that can be applied to all the different cases you encounter here. This would make the code much easier to follow and clean.

[mho291]

Hi @BrunoLiegiBastonLiegi! Most of the issues you've raised have been resolved. Thanks so much for looking through the code and providing helpful comments. The code that I have in my local device has all the resolved issues implemented and it works. Perhaps we can discuss some of the outstanding issues before I do another push. Thanks a lot!

[BrunoLiegiBastonLiegi]

Hi, thanks for addressing the comments, can you push your local changes such that I can give it a look please?

[BrunoLiegiBastonLiegi]

Thanks for the updates @mho291. I think that some improvements are possible for the GST_measure_* functions. I still have to go over the rest but I would focus on those first.

[BrunoLiegiBastonLiegi]

Thanks @mho291 for the updates. I made some further comments. In general I have the impression that this could be made a lot shorter by applying smarter design choices. Avoid hard coded definitions when possible, and try to write general helper functions that can be specialized for the different use cases. This would also make the code much easier to follow. Moreover, if there is a reference available for this implementation, please attach it to the first post in this PR.

[AlejandroSopena]

Hi @mho291, can you push your latest changes so we can continue reviewing from there please?

[mho291]

Hi @mho291, can you push your latest changes so we can continue reviewing from there please?

Just did. Sorry for the delay. I planned to push yesterday but I caught the covid bug.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (465e88d) 100.00% compared to head (2592ec8) 100.00%.

Additional details and impacted files

@@            Coverage Diff            @@
##            master     #1106   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           69        71    +2     
  Lines        10110     10202   +92     
=========================================
+ Hits         10110     10202   +92     

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[AlejandroSopena]

Thanks. The implementation of the method seems correct. Just a few comments and suggestions to shorten the code and avoid redundancies.

[AlejandroSopena]

Hi @mho291. Thanks for the changes. Here are some new suggestions:

[renatomello]

@scarrazza @andrea-pasquale @Jacfomg Following conversations with @AlejandroSopena @ QIP 2024, I'd like to propose that the validation module created here be renamed to tomography and we let this be the initial PR of a new module that would serve both for simulation of tomography (important and necessary on its own and will be necessary for future projects) and also as primitives to qibocal (e.g. https://github.com/qiboteam/qibocal/tree/main/src/qibocal/protocols/characterization/randomized_benchmarking).

[scarrazza]

Sure, the renaming makes sense.

outdated