Introduce Dynamic Definition function to the wire cutting modules.

[hitomitak]

Ported Dynamic Definition code to the wire cutting function and added its test. This function has the following features.

• Create a bin from each simulation results (dd_bin)
• Reconstruct the full probability distribute from bin (reconstruct_dd_full_distribute)

The original code used a file to perform the merge process of each result but this PR changes to do it on memory.
The following code is the example of their use.

subcircuit_instance_probabilities = evaluate_subcircuits(cuts)
dd_bins = create_dd_bin(
    subcircuit_instance_probabilities,  cuts, mem_limits=3 , recursion_depth=10 
)
prob = reconstruct_dd_full_distribution(circuit, cuts, dd_bins)
metrics, _ = verify(circuit, prob)

[coveralls]

Pull Request Test Coverage Report for Build 5482095635

• 194 of 194 (100.0%) changed or added relevant lines in 3 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+1.0%) to 91.0%

---

Totals
Change from base Build 5433896893:	1.0%
Covered Lines:	2629
Relevant Lines:	2889

---

💛 - Coveralls

[caleb-johnson]

This looks great, thanks! Looks like the linter is failing.

You should be able to

pip install black
black .

from the root of the repo. You can also run tox -elint locally (I think) to see if there are any other linter errors.

[caleb-johnson]

Really nice work, and thanks again for porting this over. We have had this as a to-do for a while now :)

Most of my comments are just pedantic docstring cleanups and commented print statement cleanups.

If you could take a look at the linter (currently failing the black --check) and also the two lines missing from coverage in the new code. If you can cover those two lines with a couple simple tests, we like to do that. If it's a pain to cover those lines for some reason, we can ignore them for coverage explicitly.

[hitomitak]

Thank you for reviewing. I should have combined your suggestions with batch into one commit. Sorry for the dirty commit log😰

I added the test to achieve 100% coverage and executed lint programs.

[caleb-johnson]

Thank you for reviewing. I should have combined your suggestions with batch into one commit. Sorry for the dirty commit log😰

I added the test to achieve 100% coverage and executed lint programs.

Hah! No worries at all! Thanks for the quick turnaround

[caleb-johnson]

I'm going to give @garrison a chance to look at this, and I'm going to make another pass shortly, but it's looking great

[garrison]

Please add the following line to the body of one of your commit messages, in order to acknowledge the original author of this code. (You could even create an empty commit, using git commit --allow-empty.)

Co-authored-by: weiT1993 <tangwei1027@gmail.com>

[hitomitak]

I added the co-author and fix lint.

[garrison]

Actually, for our purposes, I think it would be better to include the full text of the license in the file.

This PR deserves a release note, too.

[garrison]

I updated reconstruct_dd_full_distribute -> reconstruct_dd_full_distribution in 9c9164f, so it would be parallel to the existing name reconstruct_full_distribution.

[garrison]

Thank you again for this!! It will be in the v0.3 release, which we expect to tag shortly. 🎉

[caleb-johnson]

Thanks @hitomitak !

[hitomitak]

Sorry I was on summer vacation and could not respond. Thanks for the merge.