PT + ZNE tutorial #2601

purva-thakre · 2024-12-13T01:18:44Z

Description

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.

Before opening the PR, please ensure you have completed the following where appropriate.

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

purva-thakre · 2024-12-13T01:22:18Z

@natestemen Do you mind taking a quick look at the general outline of this tutorial? I have to clean up the code a bit. So, you could ignore the code blocks for now.

I am not too sure if I need to describe something in excruciating detail as the PT user guide is still WIP.

codecov · 2024-12-13T01:32:37Z

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.74%. Comparing base (0262bfc) to head (ca4b9c8).
Report is 25 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2601      +/-   ##
==========================================
+ Coverage   98.72%   98.74%   +0.01%     
==========================================
  Files          92       93       +1     
  Lines        4168     4219      +51     
==========================================
+ Hits         4115     4166      +51     
  Misses         53       53

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

natestemen · 2024-12-13T02:05:55Z

Took a quick look through things and this looks great! I think it's a nice split of theory and practice. I can imagine some of the material working its way in to the user guide, but lets not worry about that for now.

purva-thakre · 2024-12-17T16:04:42Z

@natestemen The comparison plot is missing data related to only ZNE values.

This is because the circuit I am using is very small for ZNE. If I increase the circuit size used for the comparison plot, the time taken by the tutorial during execution is a lot.

natestemen

Haven't finished reading through everything just yet, but wanted to get feedback soon so we can start iterating.

Overall this is a really great mix of theory and practice. Well done setting the bones up.

docs/source/_thumbnails/pt_zne.png

docs/source/examples/examples.md

docs/source/examples/pt_zne.md

cosenal · 2024-12-18T10:41:02Z

I restarted the RTD build; having a rendered version will make it easier to review.

natestemen

Really solid tutorial! Can't wait to get it up!

docs/source/examples/pt_zne.md

natestemen · 2025-01-15T23:05:13Z

docs/source/examples/pt_zne.md

+plt.ylabel("Expectation Values Difference")
+plt.title("Comparison of expectation values with ideal as a function of noise strength")
+plt.legend()
+plt.show()


Do we understand why ZNE is so good when the noise strength = 0.5?

I am not too sure for the circuit used in this example.

If I use a smaller GHZ circuit of 2 qubits, the noise does not do anything as the final state vector before and after adding coherent noise is the same. Interestingly, this behavior is not preserved as the number of qubits in the initial ghz circuit is increased. Worth looking into!

Co-authored-by: nate stemen <[email protected]>

natestemen

Left a few comments, but hopefully nothing major!

docs/source/examples/pt_zne.md

natestemen · 2025-01-28T14:08:18Z

docs/source/examples/pt_zne.md

+**Incoherent noise** is a process that results in the quantum system entangling with its environment i.e. this type of noise is irreversible. The system and the environment end up in a mixed state. It scales linearly in the small error limit. 
+The noise channel can be described using Pauli operators which makes it easy to analyze and simulate. Worst case error rate is directly proportional to the **average gate infidelity**. 
+
+For example, a depolarizing noise channel is a stochastic noise channel where a noiseless process is probabilistically mixed with orthogonal errors. If $\rho$ is a single qubit state, $p$ is the probabilistic error rate and $\mathcal{E}(\rho)$ is the noise channel:


This is a great example, but it needs a connection back to (in)coherent noise. Can you add something, maybe at the end?

Do you think we should have an example for coherent noise too?

The depolarizing noise example is to show how Pauli twirling is equivalent to tailoring the noise into a depolarizing noise channel.

I don't quite see how this relates to pauli twirling. Can you help spell it out?

From my perspective the main idea in pauli twirling is to add gates that compile to the identity, but change the impact of noise. So when you say the depolarizing channel can be written in terms of paulis, I don't see how that is pauli twirling.

From my perspective the main idea in pauli twirling is to add gates that compile to the identity,

Uh, no. Pauli twirling's goal is to transform some noise channel into a local depolarizing noise channel. The circuits average over to a mixed noise channel which is what I was trying to show in my example.

docs/source/examples/pt_zne.md

cosenal · 2025-01-28T15:09:15Z

docs/source/examples/pt_zne.md

+print("Example noisy twirled circuit", noisy_twirled_circuits[-1], sep="\n")
+```
+
+The twirled PTM is averaged over each noisy twirled circuit such that the new PTM is close to that of the PTM of incoherent noise. We skip the step in this section as we require a very large number of twirled circuits to demonstrate the desired effect of averaging over multiple numpy arrays.


A bit anti-climatic, I was looking forward to seeing the effect of this on those PTM heatmaps :)

Yes, well, this increases the notebook execution time by a lot. The coherent noise that I used is obviously not included in the noisy options allowed by generate_pauli_twirl_variants. We should create an issue to clean this portion of pauli twirling.

So, if I want to generate 100 pauli twirled circuits, I have to splice each of the circuits to make sure the noise is inserted in the correct spot and the pauli twirling gates are also inserted accordingly. Then, the next step is to get their PTM's and sum them up.

What if we included an image of the twirled PTM heatmap saying "this is what would happen if you generated the twirled PTM"?

I have added 3 4 examples. See 2fbf94f.

I also changed the color palette to make the heatmap easy to read.

Ideally, I would prefer this to be a gif (something like this) to show how the heatmap changes as the number of pauli twirled circuits is increased. We could keep this as a unitary hackathon issue if there are not a lot of non-quantum issues. WDYT?

I manually created a gif in 818a864 Hopefully, this fixes Alessandro's comment.

I was looking forward to seeing the effect of this on those PTM heatmaps

docs/source/examples/pt_zne.md

cosenal · 2025-01-28T15:11:49Z

docs/source/examples/pt_zne.md

+
+We are going to add coherent noise to this circuit and then get the error-mitigated expectation value. For a detailed discussion on this, refer to the [ZNE user guide](../guide/zne-1-intro.md). 
+
+As we are using a simulator, we have to make sure the noise model adds coherent noise to CZ/CNOT gates in our circuit. For this, `get_noise_model` is used to add noise to CZ/CNOT gates. See [PT user guide](../guide/pt-1-intro.md) for more. 


solid function that get_noise_model 📦 😃

docs/source/examples/pt_zne.md

Co-authored-by: nate stemen <[email protected]>

purva-thakre · 2025-01-31T21:29:20Z

@natestemen @cosenal This is ready. Whenever you guys are free...

Good luck with FOSDEM!

purva-thakre · 2025-01-31T22:22:38Z

docs/source/examples/pt_zne.md

+print("Example noisy twirled circuit", noisy_twirled_circuits[-1], sep="\n")
+```
+
+The twirled PTM is averaged over each noisy twirled circuit such that the new PTM is close to that of the PTM of incoherent noise. We skip the step in this section as we require a very large number of twirled circuits to demonstrate the desired effect of averaging over multiple numpy arrays. The variations in pauli twirled PTMs are shown below when averaged over a different number of pauli twirled circuits: [3](../img/pt_zne_3_circuits.png), [5](../img/pt_zne_5_circuits.png), [30](../img/pt_zne_30_circuits.png) and [300](../img/pt_zne_100_circuits.png).  


IDK if there's a better way to provide a link to a couple of png images. These appear as images a user will have to download.

add md file

fe9588f

add comparison plot

fc4099b

purva-thakre requested a review from natestemen December 17, 2024 15:47

purva-thakre added 2 commits December 17, 2024 09:50

add thumbnail image

e01d362

correct plot typo

083d254

natestemen requested a review from cosenal December 17, 2024 16:55

natestemen reviewed Dec 17, 2024

View reviewed changes

cosenal reviewed Dec 18, 2024

View reviewed changes

purva-thakre added 2 commits December 18, 2024 11:46

update plot

d07e630

nate/alessandro feedback

cf8b1a5

purva-thakre force-pushed the pt_zne_tutorial branch from 3f6f478 to cf8b1a5 Compare December 20, 2024 14:41

correct typo

b1f04de

purva-thakre requested review from natestemen and cosenal December 20, 2024 14:53

purva-thakre mentioned this pull request Dec 20, 2024

LRE failure for chunking #2608

Open

6 tasks

natestemen requested changes Jan 15, 2025

View reviewed changes

purva-thakre and others added 2 commits January 15, 2025 19:37

Apply suggestions from code review

5e59346

Co-authored-by: nate stemen <[email protected]>

check build: set heatmap bounds and y axis label

d0d4b2c

purva-thakre force-pushed the pt_zne_tutorial branch from 5c95219 to d0d4b2c Compare January 16, 2025 04:12

nate's suggestions to code, update thumbnail image

ac50eb8

purva-thakre requested a review from natestemen January 23, 2025 18:31

ordering phrasing

bba28b6

natestemen reviewed Jan 28, 2025

View reviewed changes

cosenal reviewed Jan 28, 2025

View reviewed changes

some suggested changes and example twirled ptms

2fbf94f

purva-thakre and others added 3 commits January 30, 2025 18:08

Update docs/source/examples/pt_zne.md

d3046d7

Co-authored-by: nate stemen <[email protected]>

remaining suggestions

8f3a08d

add 100 twirled circuits plot

c890f2b

purva-thakre force-pushed the pt_zne_tutorial branch from 4ea318f to c890f2b Compare January 31, 2025 20:35

purva-thakre requested review from natestemen and cosenal January 31, 2025 21:19

typo lets to let's

6f3cd94

change to gif

818a864

purva-thakre commented Jan 31, 2025

View reviewed changes

typo 300 to 100 circuits

ca4b9c8

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

PT + ZNE tutorial #2601

PT + ZNE tutorial #2601

purva-thakre commented Dec 13, 2024

purva-thakre commented Dec 13, 2024

codecov bot commented Dec 13, 2024 •

edited

Loading

natestemen commented Dec 13, 2024

purva-thakre commented Dec 17, 2024

natestemen left a comment

cosenal commented Dec 18, 2024

natestemen left a comment

natestemen Jan 15, 2025

purva-thakre Jan 16, 2025

natestemen left a comment

natestemen Jan 28, 2025

purva-thakre Jan 28, 2025

natestemen Jan 28, 2025

purva-thakre Jan 30, 2025 •

edited

Loading

cosenal Jan 28, 2025

purva-thakre Jan 28, 2025

natestemen Jan 28, 2025 •

edited

Loading

purva-thakre Jan 30, 2025 •

edited

Loading

cosenal Jan 28, 2025

purva-thakre commented Jan 31, 2025

purva-thakre Jan 31, 2025


		We are going to add coherent noise to this circuit and then get the error-mitigated expectation value. For a detailed discussion on this, refer to the [ZNE user guide](../guide/zne-1-intro.md).

		As we are using a simulator, we have to make sure the noise model adds coherent noise to CZ/CNOT gates in our circuit. For this, `get_noise_model` is used to add noise to CZ/CNOT gates. See [PT user guide](../guide/pt-1-intro.md) for more.

PT + ZNE tutorial #2601

Are you sure you want to change the base?

PT + ZNE tutorial #2601

Conversation

purva-thakre commented Dec 13, 2024

Description

License

purva-thakre commented Dec 13, 2024

codecov bot commented Dec 13, 2024 • edited Loading

Codecov Report

natestemen commented Dec 13, 2024

purva-thakre commented Dec 17, 2024

natestemen left a comment

Choose a reason for hiding this comment

cosenal commented Dec 18, 2024

natestemen left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

natestemen left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

purva-thakre Jan 30, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

natestemen Jan 28, 2025 • edited Loading

Choose a reason for hiding this comment

purva-thakre Jan 30, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

purva-thakre commented Jan 31, 2025

Choose a reason for hiding this comment

codecov bot commented Dec 13, 2024 •

edited

Loading

purva-thakre Jan 30, 2025 •

edited

Loading

natestemen Jan 28, 2025 •

edited

Loading

purva-thakre Jan 30, 2025 •

edited

Loading