Create visualization tests for pulse

[enavarro51]

What is the expected enhancement?

As part of #6467, some old pulse image comparison tests that were being skipped were removed. These were in the file test_pulse_visualization_output.txt. The tests in this file may be useful when a #4544 binder style set of image comparison tests is created for pulse.

[MartinBeseda]

Dear @enavarro51 ,
Dear @taalexander ,

I'd have a look at this, if possible.

[javabster]

hey @MartinBeseda sorry for the slow response, are you still interested in working on this? If yes lmk and I'll assign to you 😄

[ozamanan]

Hey, @javabster can I work on this?

[javabster]

Hi @ozamanan as @MartinBeseda is unresponsive I will assign to you, let us know if you have any questions 😄

[anonymousr007]

Hi @javabster, Add text_pulse_visualization_output.txt in this folder /test/python/pulse. Right ?

[javabster]

Hi @javabster, Add text_pulse_visualization_output.txt in this folder /test/python/pulse. Right ?

Hi @anonymousr007 actually I think the point of this issue is to create some binder style snapshot tests for pulse, using the tests found in text_pulse_visualization_output.txt. The other binder tests can be found in test/ipynb, I would suggest creating a subfolder test/ipynb/pulse and following the same format that is used for the other circuit and graph tests.

[1ucian0]

Any news @ozamanan ?

[1ucian0]

Hi @anonymousr007
The assigned contributor is unresponsive are you interested in taking over?

[AbdullahKazi500]

The expected enhancement is to create a set of image comparison tests for pulse, similar to the existing #4544 binder style tests. These tests will help ensure that the pulse images generated by different pulse programs are consistent and correct. The tests can be created based on the old pulse image comparison tests that were previously skipped, which were removed as part of #6467. By adding these tests back in a new form, the pulse functionality can be more thoroughly tested and validated @1ucian0 Can I Get assigned to this issue

[1ucian0]

Thanks @AbdullahKazi500 ! assigning.

Small headsup. The binder mechanism is growing intro proper CI testing in #9961 . There are two options at this point:

• You can write pulse tests top of AngeloDanducci:ad9892. This creates a dependency and a potential blocker, but it does not need follow up.
• You can do it in the notebook way and then we moved them to the CI way. This allows you to work more independently, but it might requiere an update if 9961 lands before yours.

Up to you. Whatever you feel more confortable with.

[AbdullahKazi500]

Thank you for the heads up! It's good to know that the binder mechanism is moving towards proper CI testing. I'll keep these options in mind when writing pulse tests. It seems like using the notebook way first and then moving to CI testing would be the safer option in case there are any changes in #9961.

[Raghav-Bell]

Hi @1ucian0! It seems #9961 is merged, binder tests are integrated in the workflow. Is this issue still open ?

[AbdullahKazi500]

Hi @Raghav-Bell Yeah I am working on this for now

[1ucian0]

Hi @AbdullahKazi500 How is that going?

[AbdullahKazi500]

@1ucian0 Hi Luciano I have created a Jupyter Notebook-based tests for pulse that run as part of the Continuous Integration (CI) process. These notebooks should include the necessary test cases and comparisons.
Once I have implemented the initial CI tests, I will try to create a new branch in your fork of the Qiskit repository,
I am trying to Locate the Tests and Review the contents of the file test_pulse_visualization_output.txt in the Qiskit repository to understand the older image comparison tests that were removed.
whether these tests are still relevant and whether they can be incorporated into the new CI testing framework for pulse.
Depending on their relevance and the current state of the code, I Can reimplement these tests or adapt them to the new testing framework.
If necessary, I can create new tests based on the older tests.