Add probabilities to MeasurementOutcomes

[BrunoLiegiBastonLiegi]

Tiny patch to #1039 that implements the calculation of the probabilities starting from the frequencies in MeasurementOutcomes objects.

Checklist:

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

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (b5a6a2c) 100.00% compared to head (60f4253) 100.00%.

Additional details and impacted files

@@            Coverage Diff            @@
##            master     #1066   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           63        63           
  Lines         8835      8857   +22     
=========================================
+ Hits          8835      8857   +22     

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.

[scarrazza]

@BrunoLiegiBastonLiegi did you test this on hardware?

[BrunoLiegiBastonLiegi]

Not yet, I am going to test it asap.

[stavros11]

Thanks @BrunoLiegiBastonLiegi, looks good to me. The comments are not really blocking, although I would give a try to lower the tolerance in the test.

I also have not tested on hardware (will test the qibolab PR soon), but I don't the current PR is breaking anything since you are just calculating the probabilities from frequencies, and it shouldn't make a difference if the frequencies are simulated or obtained from hardware.

[BrunoLiegiBastonLiegi]

Ok, I made a quick test on the hardware and the calculation of the probabilities from the frequencies works.
I'll do another one after I added the qubits parameter to probabilities here as well.
One thing it's not completely clear to me though. If the user asks for a subset of the measured qubits, should you renormalize the probabilities of the selected qubits such that they still sum to one, or you just ignore the other qubits and return the global probabilities of the selected ones?

[BrunoLiegiBastonLiegi]

Ok I have added the qubits parameter and tested on hardware. It seems to work fine up to the calibration of the qubits.

Outdated