noise model done by eshant742 · Pull Request #14 · adelshb/graphqec

eshant742 · 2025-02-13T13:47:06Z

No description provided.

adelshb · 2025-02-13T18:55:38Z

Hi @eshant742, thank you for the PR!
I went through a few errors while running the tests. Could you have a look and proceed with the changes?

NoiseModel import issues: Adding "from .noise_model import * # noqa" in the graphqec/init.py file
Change all the imports from "from graphqec.noise_models" to "from graphqec.noise_model" (without the extra "s" at model)
stim.Circuit have no method 'to_text_diagram' (Stim API reference: https://github.com/quantumlib/Stim/blob/main/doc/python_api_reference_vDev.md)

Sorry to ask you for extra effort. Let me know if there is anything unclear!

eshant742 · 2025-02-14T06:53:35Z

yeah i am on it

eshant742 · 2025-02-14T07:10:23Z

these were changes
NoiseModel Import

Added from .noise_model import * # noqa to graphqec/init.py so that NoiseModel and related classes are accessible without import errors.
Changed all imports from from graphqec.noise_models to from graphqec.noise_model (removing the extra "s") to match the actual file name.
Stim to_text_diagram

Replaced circuit.to_text_diagram() with str(circuit) because to_text_diagram() is only available in Stim’s development branch. Using str(circuit) works on the stable version of Stim, preventing the “no method ‘to_text_diagram’” error.

adelshb · 2025-02-15T14:24:19Z

graphqec/lab/threshold/threshold_lab.py

 from graphqec.codes.base_code import BaseCode

+# Import the NoiseModel interface (and possibly other models) if needed
+from graphqec.noise_models import NoiseModel


The ending "s" is still in the import

The "s" is still there

adelshb · 2025-02-15T18:04:56Z

graphqec/lab/threshold/threshold_lab.py

    def __init__(
-        self, code: BaseCode, distances: list[int], error_rates: list[float]
+        self,
+        code: BaseCode,        # or `type[BaseCode]` if `code` is a class, not an instance


Replace with type[BaseCode]
(Thank you for pointing it out!)

adelshb · 2025-02-15T18:05:14Z

graphqec/lab/threshold/threshold_lab.py

 import matplotlib.pyplot as plt
 import pymatching

+# If your code parameter is actually a class (subclass of BaseCode),


Remove comment

adelshb · 2025-02-15T18:08:20Z

graphqec/__init__.py

 from .measurement import *  # noqa
 from .stab import *  # noqa
 from .lab import *  # noqa
+# graphqec/__init__.py


Comment to be removed

adelshb · 2025-02-15T18:08:29Z

graphqec/codes/base_code.py

 from graphqec.measurement import Measurement
 from graphqec.stab import X_check, Z_check

+# NEW: Import your NoiseModel and a default DepolarizingNoiseModel


Comment to be removed

adelshb · 2025-02-15T18:08:33Z

graphqec/codes/base_code.py

@@ -1,3 +1,5 @@
+# graphqec/codes/base_code.py


Comment to be removed

adelshb · 2025-02-15T18:08:42Z

graphqec/codes/base_code.py

        "_graph",
        "_checks",
        "_logic_check",
+        "_noise_model",  # NEW SLOT


Comment to be removed

adelshb · 2025-02-15T18:08:46Z

graphqec/codes/base_code.py

        distance: int = 3,
        depolarize1_rate: float = 0,
        depolarize2_rate: float = 0,
+        noise_model: NoiseModel = None,  # NEW PARAM


Comment to be removed

adelshb · 2025-02-15T18:22:20Z

graphqec/codes/base_code.py

-        self._checks: list[str]
-        self._logic_check: list[str]
-
+        self._checks: list[str] = []


To be removed. It should be defined in the Child class

adelshb · 2025-02-15T18:22:24Z

graphqec/codes/base_code.py

-        self._logic_check: list[str]
-
+        self._checks: list[str] = []
+        self._logic_check: list[str] = []


To be removed. It should be defined in the Child class

adelshb · 2025-02-15T18:41:02Z

There is an issue with the implementation. The threshold lab returns 0s on the collected data. For the following code:

th = ThresholdLAB(
    distances= [3, 5],
    code=RepetitionCode,
    error_rates= np.linspace(0, 0.1, 10)
)

noise = DepolarizingNoiseModel()
th.collect_stats(num_shots=10**3, noise_model=noise)

print(th.collected_stats)

I get:

{3: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
 5: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]}

we should expect non-zero values for non-zero error rates. Could you have a look?

noise model done

0827505

noise model done

af4cb56

adelshb reviewed Feb 15, 2025

View reviewed changes

updating

a6eb910

adelshb reviewed Feb 15, 2025

View reviewed changes

graphqec/codes/base_code.py

@@ -1,3 +1,5 @@

# graphqec/codes/base_code.py

Copy link

Owner

adelshb Feb 15, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Comment to be removed

adelshb reviewed Feb 15, 2025

View reviewed changes

graphqec/codes/base_code.py

"_graph",

"_checks",

"_logic_check",

"_noise_model", # NEW SLOT

Copy link

Owner

adelshb Feb 15, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Comment to be removed

adelshb reviewed Feb 15, 2025

View reviewed changes

Conversation

eshant742 commented Feb 13, 2025

Uh oh!

adelshb commented Feb 13, 2025

Uh oh!

eshant742 commented Feb 14, 2025

Uh oh!

eshant742 commented Feb 14, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adelshb commented Feb 15, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

adelshb commented Feb 15, 2025 •

edited

Loading