[unitaryhack] Updated molecular hydrogen tutorial #1296

lockwo · 2022-05-17T16:39:14Z

Description

This is made to address #1094. The goal is to address both components of that issue. I made it so the hamiltonian is calculated using openfermion. The hamiltonian is not the same as in the paper (still 2 qubits, just not the same ops). It would be very easy to programmatically change that, I just don't know how to do that exact symmetry reduction in openfermion. Additionally, the hamiltonian is a mitiq.Observable and the executor is no longer a wrapper for the energy function but just the density matrix simulator. I am new to mitiq so this second part is what I think that issue meant, but if it should change just let me know.

Checklist

Check off the following once complete (or if not applicable) after opening the PR. The PR will be reviewed once this checklist is complete and all tests are passing.

I added unit tests for new code.
I used type hints in function signatures.
I used Google-style docstrings for functions.
I updated the documentation where relevant.
Added myself / the copyright holder to the AUTHORS file

If some items remain, you can mark this a draft pull request.

License

I license this contribution under the terms of the GNU GPL, version 3 and grant Unitary Fund the right to provide additional permissions as described in section 7 of the GNU GPL, version 3.

Tips

If the validation check fails:
1. Run make check-types (from the root directory of the repository) and fix any mypy errors.
2. Run make check-style and fix any flake8 errors.
3. Run make format to format your code with the black autoformatter.
For more information, check the Mitiq style guidelines.
Write "Fixes #XYZ" in the description if this PR fixes Issue #XYZ.

github-actions · 2022-05-17T16:39:28Z

👈 Launch a binder notebook on branch lockwo/mitiq/master

github-actions

Hello @lockwo, thank you for submitting a PR to Mitiq! We will respond as soon as possible, and if you have any questions in the meantime, you can ask us on the Unitary Fund Discord.

codecov · 2022-05-17T17:03:25Z

Codecov Report

Merging #1296 (15340e1) into master (0440b16) will increase coverage by 0.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1296      +/-   ##
==========================================
+ Coverage   98.13%   98.15%   +0.01%     
==========================================
  Files          60       61       +1     
  Lines        2739     2762      +23     
==========================================
+ Hits         2688     2711      +23     
  Misses         51       51

Impacted Files	Coverage Δ
mitiq/benchmarks/__init__.py	`100.00% <0.00%> (ø)`
mitiq/benchmarks/quantum_volume_circuits.py	`100.00% <0.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0440b16...15340e1. Read the comment docs.

lockwo · 2022-05-17T17:18:42Z

I think the check is unsuccessful (Couldn't find cache key for notebook file source/examples/molecular_hydrogen.myst) because OpenFermion isn't installed (based on executablebooks/jupyter-cache#51) since it builds locally.

andreamari · 2022-05-18T08:34:25Z

Thanks @lockwo, very interesting!
You can try to add openfermion in https://github.com/unitaryfund/mitiq/blob/master/dev_requirements.txt under the documentation and examples section.
Not sure if it works (especially with windows tests) but worth trying.

As a second option, we can try to add !pip install openfermion directly on the notebook.

nathanshammah · 2022-05-18T10:10:25Z

Thanks @lockwo. I like the fact that the hard coded table is replaced with values taken from generate_molecular_hamiltonian.

It would be nice to add a bit of details about this function and its parameters.
It would be nice to keep a reference of {cite}OMalley_2016_PRX, even if now the data is not directly taken from there.

lockwo · 2022-05-19T00:53:15Z

Thanks @lockwo. I like the fact that the hard coded table is replaced with values taken from generate_molecular_hamiltonian.

It would be nice to add a bit of details about this function and its parameters.

It would be nice to keep a reference of {cite}OMalley_2016_PRX, even if now the data is not directly taken from there.

I expanded a bit on the function description. Is there anything specific you would like to see? As for the citation, I still see it there (in the image). Did you want that referenced again when talking about the coefficients?

lockwo · 2022-05-19T03:06:21Z

Thanks @lockwo, very interesting! You can try to add openfermion in https://github.com/unitaryfund/mitiq/blob/master/dev_requirements.txt under the documentation and examples section. Not sure if it works (especially with windows tests) but worth trying.

As a second option, we can try to add !pip install openfermion directly on the notebook.

Great, this first suggestion worked!

andreamari · 2022-05-19T10:52:40Z

Hi @lockwo,
as you mentioned, the Hamiltonian produced by openfermion is probably different from the formula visible in the comment above.
Any idea to fix this inconsistency?
Some ideas from my side:

Change the formula to something more generic e.g. a sum over generic Pauli operators.
Change the formula to something exactly matching the operators produced by openfermion
Keep the previous table of parameters and add openfermion as a "second option" to generate the Hamiltonian.
Other options are welcome.

lockwo · 2022-05-19T14:51:36Z

Hi @lockwo, as you mentioned, the Hamiltonian produced by openfermion is probably different from the formula visible in the comment above. Any idea to fix this inconsistency? Some ideas from my side:

Change the formula to something more generic e.g. a sum over generic Pauli operators.

Change the formula to something exactly matching the operators produced by openfermion

Keep the previous table of parameters and add openfermion as a "second option" to generate the Hamiltonian.

Other options are welcome.

So the formula is the similar it just doesn't have the YY term (hence why the unmitigated case seems to do a tiny bit better it looks), the symmetry reduction comes from: https://arxiv.org/abs/1701.08213. However, this also makes it so the circuit doesn't need the initial X gate on the second qubit. I think a combination of your suggestions might work. Perhaps adding the exact description of the OF Hamiltonian in addition to providing the parameters from the original work might be a good way to do it. I will make those changes in a little bit to see how they look.

andreamari · 2022-05-19T15:31:16Z

Sounds good! Considering that unitaryhack will start in two weeks there is no rush.

lockwo · 2022-05-29T00:16:27Z

Alright, finally got around to making those changes.

andreamari · 2022-06-02T09:55:51Z

Thanks, @lockwo. Next week I'll make a proper review, but looks already very good!

lockwo · 2022-06-04T00:35:40Z

Sounds good!

andreamari

Thanks @lockwo, I added some comments.
Please have a look when you have some time.
This is close to be accepted.
Very nice work.

docs/source/examples/molecular_hydrogen.myst

andreamari · 2022-06-07T15:10:11Z

docs/source/examples/molecular_hydrogen.myst

+    qubit_operator = qubit_operator_to_pauli_sum(bk_op)
+    return Observable(*qubit_operator)
+
+hamiltonians = [get_hamiltonian(i) for i in radii]


Is there any simple way to show that these Hemiltonians have indeed the structure mentioned in the latex formula?
E.g. something like

print(get_hamiltonian( some_both_lenght)).

Or any other way to check which Hamiltonian OpenFermion is actually producing?

I added a print function for the hamiltonian, so you can see it is of the same form. Let me know if that is what you were looking for.

nathanshammah · 2022-06-08T16:47:00Z

+1 this looks very neat, especially for the use of OpenFermion for the tabulated data instead of a hardcoded table.

Co-authored-by: Andrea Mari <[email protected]>

nathanshammah · 2022-06-08T21:32:45Z

Can you add the use of OpenFermion in the notebook title? To differentiate from #1337 that is done with PennyLane.

lockwo · 2022-06-09T03:46:38Z

Hmmm it has somehow broken the build, it works when I build locally. I see an error with the QAOA tutorial, but I'm not sure if that is the source (since this link is now dead: https://quantumai.google/cirq/tutorials/qaoa) I will fix it anyway and see

lockwo · 2022-06-09T04:11:18Z

That was the problem actually (not sure if you want to make that fix a separate PR, but it is in this one).

lockwo · 2022-06-09T04:11:30Z

Can you add the use of OpenFermion in the notebook title? To differentiate from #1337 that is done with PennyLane.

Done

andreamari

Thank you @lockwo, LGTM!
And thanks for fixing the broken link too!

Moreover congratulations for merging a unitaryhack PR and so for getting the associated bounty!

This reverts commit fe58bb2.

* Revert "[unitaryhack] Updated molecular hydrogen tutorial (#1296)" This reverts commit fe58bb2. * fix broken link

andreamari · 2022-06-09T17:00:44Z

Hi @lockwo, there was a problem with OpenFermion not working with Windows tests on the master branch. So we had to temporarily revert the merging commit.

Of course we still want to have your example but we need to merge it more carefully. Could you open a new PR with the same code as this one? Re-opening this one is fine too (if possible).

lockwo · 2022-06-09T17:12:46Z

Just made it

lockwo added 3 commits May 17, 2022 12:27

Update molecular_hydrogen.myst

6322067

Update molecular_hydrogen.myst

ed29f02

Update AUTHORS

c8e0b01

github-actions bot reviewed May 17, 2022

View reviewed changes

lockwo added 2 commits May 18, 2022 20:50

Update molecular_hydrogen.myst

7f8b053

Update dev_requirements.txt

2705630

lockwo added 2 commits May 28, 2022 13:43

Update molecular_hydrogen.myst

06829b8

Merge branch 'master' into master

8984c80

lockwo marked this pull request as ready for review May 30, 2022 19:37

lockwo changed the title ~~Updated molecular hydrogen tutorial [draft] [unitaryhack]~~ [unitaryhack] Updated molecular hydrogen tutorial Jun 2, 2022

andreamari reviewed Jun 7, 2022

View reviewed changes

andreamari linked an issue Jun 7, 2022 that may be closed by this pull request

[unitaryhack-bounty] Improve H2 example #1094

Closed

2 tasks

lockwo and others added 2 commits June 8, 2022 15:03

Update docs/source/examples/molecular_hydrogen.myst

39682ef

Co-authored-by: Andrea Mari <[email protected]>

Update molecular_hydrogen.myst

f23df44

Update molecular_hydrogen.myst

6ddfa84

Update maxcut-demo.myst

15340e1

andreamari approved these changes Jun 9, 2022

View reviewed changes

andreamari added the unitaryhack-accepted label Jun 9, 2022

andreamari merged commit fe58bb2 into unitaryfund:master Jun 9, 2022

andreamari added a commit that referenced this pull request Jun 9, 2022

Revert "[unitaryhack] Updated molecular hydrogen tutorial (#1296)"

4a6356b

This reverts commit fe58bb2.

andreamari mentioned this pull request Jun 9, 2022

Revert "[unitaryhack] Updated molecular hydrogen tutorial" #1347

Merged

andreamari added a commit that referenced this pull request Jun 9, 2022

Revert "[unitaryhack] Updated molecular hydrogen tutorial" (#1347)

7f8fee0

* Revert "[unitaryhack] Updated molecular hydrogen tutorial (#1296)" This reverts commit fe58bb2. * fix broken link

lockwo mentioned this pull request Jun 9, 2022

[unitaryhack] v2 Updated molecular hydrogen tutorial OpenFermion #1349

Merged

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[unitaryhack] Updated molecular hydrogen tutorial #1296

[unitaryhack] Updated molecular hydrogen tutorial #1296

lockwo commented May 17, 2022

github-actions bot commented May 17, 2022

github-actions bot left a comment

codecov bot commented May 17, 2022 •

edited

Loading

lockwo commented May 17, 2022 •

edited

Loading

andreamari commented May 18, 2022

nathanshammah commented May 18, 2022

lockwo commented May 19, 2022

lockwo commented May 19, 2022

andreamari commented May 19, 2022

lockwo commented May 19, 2022

andreamari commented May 19, 2022

lockwo commented May 29, 2022

andreamari commented Jun 2, 2022

lockwo commented Jun 4, 2022

andreamari left a comment

andreamari Jun 7, 2022 •

edited

Loading

lockwo Jun 8, 2022

nathanshammah commented Jun 8, 2022

nathanshammah commented Jun 8, 2022 •

edited

Loading

lockwo commented Jun 9, 2022

lockwo commented Jun 9, 2022

lockwo commented Jun 9, 2022

andreamari left a comment

andreamari commented Jun 9, 2022

lockwo commented Jun 9, 2022

[unitaryhack] Updated molecular hydrogen tutorial #1296

[unitaryhack] Updated molecular hydrogen tutorial #1296

Conversation

lockwo commented May 17, 2022

Description

Checklist

License

Tips

github-actions bot commented May 17, 2022

github-actions bot left a comment

Choose a reason for hiding this comment

codecov bot commented May 17, 2022 • edited Loading

Codecov Report

lockwo commented May 17, 2022 • edited Loading

andreamari commented May 18, 2022

nathanshammah commented May 18, 2022

lockwo commented May 19, 2022

lockwo commented May 19, 2022

andreamari commented May 19, 2022

lockwo commented May 19, 2022

andreamari commented May 19, 2022

lockwo commented May 29, 2022

andreamari commented Jun 2, 2022

lockwo commented Jun 4, 2022

andreamari left a comment

Choose a reason for hiding this comment

andreamari Jun 7, 2022 • edited Loading

Choose a reason for hiding this comment

lockwo Jun 8, 2022

Choose a reason for hiding this comment

nathanshammah commented Jun 8, 2022

nathanshammah commented Jun 8, 2022 • edited Loading

lockwo commented Jun 9, 2022

lockwo commented Jun 9, 2022

lockwo commented Jun 9, 2022

andreamari left a comment

Choose a reason for hiding this comment

andreamari commented Jun 9, 2022

lockwo commented Jun 9, 2022

codecov bot commented May 17, 2022 •

edited

Loading

lockwo commented May 17, 2022 •

edited

Loading

andreamari Jun 7, 2022 •

edited

Loading

nathanshammah commented Jun 8, 2022 •

edited

Loading