Fixes to various graphs in the docs

[gadial]

Summary

Fixes visual problems in the display of some of the visualizations in the docs.

Fixes #13558
Fixes #13632

Details and comments

What's fixed

• https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.GraphState: Problem with the (obsolete) Graph state circuits using a name that includes the whole adjacency matrix. Fix by explicitly setting name.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.QuantumVolume: The graph displaying the global phase messed up the subplots. Fixed by setting the global phase to 0 when drawing. Also changed the positioning of the table to “top” to make sure there’s no gap between it and the circuit.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.InnerProduct: The circuit is missing space since drawing assumes standard size but the height is only half. This is fixed by increasing the size available to the subplot.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.transpiler.passes.DynamicalDecoupling#dynamicaldecoupling: This results from padding not playing well with sphinx. Doing self.figure.tight_layout() when calling get_image fixes this.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.visualization.plot_histogram#qiskitvisualizationplot_histogram: Similar to the one above - adding tight_layout when returning the figure fixes the way sphinx displays it.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.visualization.plot_bloch_multivector#qiskitvisualizationplot_bloch_multivector: This is caused by adding the title in position y=1.0 + title_pad / 100 which is greater than 1; changing to 0.98 and enlarging the height of the image instead fixes this.
• https://docs.quantum.ibm.com/api/qiskit/qiskit.visualization.plot_state_paulivec#qiskitvisualizationplot_state_paulivec: Also solved by adding tight_layout

What was not fixed

• https://docs.quantum.ibm.com/api/qiskit/pulse#pulse-builder: As we are removing pulse anyway (should I still do it?)
• https://docs.quantum.ibm.com/api/qiskit/qiskit.visualization.timeline_drawer#qiskitvisualizationtimeline_drawer: This is not a simple problem. Basically, there’s simply not enough space in the slice to display its name. We either need to modify the example to ensure the slice is larger, or we need some overhaul of the timeline drawer that can somehow stretch slices while keeping everything in the correct size.

Before/after screenshots

GratphState

QuantumVolume

InnerProduct

DynamicalDecoupling

[qiskit-bot]

One or more of the following people are relevant to this code:

• @enavarro51
• @Cryoris
• @Qiskit/terra-core
• @ajavadia
• @nkanazawa1989

[ShellyGarion]

can you also fix this similar issue? #13632

[ShellyGarion]

Please also verify the circuit library updated classes:
https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.GraphStateGate
https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.quantum_volume
https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.InnerProductGate

[gadial]

can you also fix this similar issue? #13632

Already handled by this PR.

Please also verify the circuit library updated classes: https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.GraphStateGate https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.quantum_volume https://docs.quantum.ibm.com/api/qiskit/qiskit.circuit.library.InnerProductGate

All look good.

[coveralls]

Pull Request Test Coverage Report for Build 14019415716

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 6 of 22 (27.27%) changed or added relevant lines in 4 files are covered.
• 67 unchanged lines in 9 files lost coverage.
• Overall coverage decreased (-0.02%) to 88.06%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
qiskit/visualization/counts_visualization.py	2	4	50.0%
qiskit/visualization/library.py	0	3	0.0%
qiskit/visualization/timeline/plotters/matplotlib.py	0	4	0.0%
qiskit/visualization/state_visualization.py	4	11	36.36%

Files with Coverage Reduction	New Missed Lines	%
crates/accelerate/src/unitary_synthesis.rs	1	94.79%
crates/qasm2/src/expr.rs	1	94.23%
qiskit/visualization/state_visualization.py	1	57.37%
qiskit/visualization/timeline/plotters/matplotlib.py	1	0.0%
qiskit/circuit/library/standard_gates/u1.py	3	94.57%
crates/qasm2/src/lex.rs	7	91.98%
crates/circuit/src/packed_instruction.rs	8	94.23%
crates/circuit/src/operations.rs	11	91.0%
crates/accelerate/src/two_qubit_decompose.rs	34	91.67%

Totals
Change from base Build 13949351916:	-0.02%
Covered Lines:	72616
Relevant Lines:	82462

---

💛 - Coveralls

[ElePT]

Thanks @gadial. Given that the fix only affects docs generated by sphinx, I think we can consider this a documentation fix, which doesn't require a release note.

[ElePT]

Thanks @gadial! The changes LGTM. The visual test references are always very annoying, because depending on the OS and other small factors, the images show this small discrepancy that breaks tests. I think that reducing the tolerance is ok in this case, not perfect, but comparing pixels is not a perfect method anyways. Alternatively, what usually do in these cases, is that I copy the output image generated in the CI visual tests published artifacts (here), but I wouldn't bother doing it in this case, I can see in the artifacts that the images are essentially correct.