Add a new gauge for SqrtCZ and support SqrtCZ† and fix and improve spin inversion gauge #6571
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #6571 +/- ##
==========================================
- Coverage 97.79% 97.78% -0.01%
==========================================
Files 1124 1124
Lines 95659 95699 +40
==========================================
+ Hits 93550 93580 +30
- Misses 2109 2119 +10 ☔ View full report in Codecov by Sentry. |
NoureldinYosri
changed the title
| return 5.0 | ||
|
|
||
| def sample(self, gate: ops.Gate, prng: np.random.Generator) -> ConstantGauge: | ||
| if prng.choice([True, False]): |
There was a problem hiding this comment.
If this can be called many times, consider using prng.integers(2) here and at other places.
prng.integers is about 5 times faster as it does not deal with a throw-away list.
NoureldinYosri
changed the title
|
@pavoljuhas when I use |
This seems to happen because Do we need to import If optional, we may save ourselves some headaches with circular imports or strange typechecking errors down the line. |
| class TestSpinInversionGauge_3(GaugeTester): | ||
| two_qubit_gate = cirq.ZZ**0.3 | ||
| gauge_transformer = SpinInversionGaugeTransformer |
There was a problem hiding this comment.
Is this test qualitatively different (ie testing different code paths) from TestSpinInversionGauge_1?
If not please remove.
There was a problem hiding this comment.
this is a correctness test, making sure that the transformer works for all rotation angles. The reason that I create several is that I can't do pytest.mark.parametrize on the test class
this transfomer should work on Ising models where the strength of two qubit interactions g Z_i Z_j are encoded in the angels of ZZ(q[i], q[j])**g
| post_q1=X, | ||
| post_q0=S if gate == _SQRT_CZ else _ADJ_S, | ||
| two_qubit_gate=gate**-1, | ||
| swap_qubits=True, |
There was a problem hiding this comment.
A bit perplexed - the swap_qubits should have no effect as the gate**-1 will be symmetric for the q0, q1 swap.
@eliottrosenberg - is this functionality for simulation / execution of circuits that may have imperfect (ie asymmetric), SQRT_CZ, INV_SQRT_CZ gates?
There was a problem hiding this comment.
gauge compiling is done because the two qubit gate is assumed to be imperfect
…in inversion gauge (quantumlib#6571)
…in inversion gauge (quantumlib#6571)
…in inversion gauge (quantumlib#6571)
No description provided.