incorrect results graph in Advanced techniques for QAOA
URL to the relevant tutorial
basically the same issue as #3523 but in a different page. page url: on https://quantum.cloud.ibm.com/docs/en/tutorials/advanced-techniques-for-qaoa section link: https://quantum.cloud.ibm.com/docs/en/tutorials/advanced-techniques-for-qaoa#step-4-post-process-and-return-result-in-desired-classical-format
This image
seems to have the colors' order reversed (Qiskit order VS python order).
It is easy to see that this is far from optimal as it contains linear sequences of same-colored nodes and singly connected nodes with the same color as their neighbor.
An approximate count found about 20+23 same color pairs and only 61 different color pairs in the graph, while the reported value of the cut is 83
Select all that apply
- [ ] new content request
- [ ] typo
- [x] code bug
- [ ] out-of-date content
- [ ] broken link
- [ ] other
Describe the fix or the content request.
rerun the code with reverse on the bit-string vector prior to coloring graph, save result and publish.
For new content requests - if the request is accepted, do you want to write the content?
I will write (or already have written) a draft of the proposed content
@a-matsuo
We're going to put this on hold since the SwapStrategy transpiler pass is hanging when trying to run the notebook. There's an issue in rustworkx tracking the problem: https://github.com/Qiskit/rustworkx/issues/1514
I think #4315 may close this too. Let's check the correctness of the graph
addressed by #4315