Allow users to run readout benchmarking with sweep #7435
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #7435 +/- ##
========================================
Coverage 98.69% 98.69%
========================================
Files 1112 1112
Lines 98076 98199 +123
========================================
+ Hits 96798 96921 +123
Misses 1278 1278 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
cc: @eliottrosenberg this is a smaller RP I split from #7358. This one only touches the readout benchmarking but not pauli expectation calculation. |
| @@ -84,6 +101,79 @@ def _generate_readout_calibration_circuits( | |||
| return readout_calibration_circuits, random_bitstrings | |||
|
|
|||
|
|
|||
| def _generate_parameterized_readout_calibration_circuit_with_sweep( | |||
| qubits: list[ops.Qid], rng: np.random.Generator, num_random_bitstrings: int | |||
There was a problem hiding this comment.
traditionally rng should be the last input, same below
| qubits_set: set[ops.Qid] = set() | ||
| for circuit in input_circuits: | ||
| qubits_set.update(circuit.all_qubits()) | ||
| qubits_to_measure = [sorted(qubits_set)] |
There was a problem hiding this comment.
nit: this can be a single line
| qubits_to_measure = [sorted(qubits_set)] | |
| qubits_to_measure = [sorted(set(q for circuit in input_circuits for q in circuit.all_qubits())] |
cirq-core/cirq/contrib/shuffle_circuits/shuffle_circuits_with_readout_benchmarking.py
Show resolved
Hide resolved
cirq-core/cirq/contrib/shuffle_circuits/shuffle_circuits_with_readout_benchmarking.py
Show resolved
Hide resolved
| # Check readout_repetitions is bigger than 0 | ||
| if readout_repetitions <= 0: | ||
| raise ValueError("Must provide non-zero readout_repetitions for readout calibration.") | ||
| def _validate_experiment_input_with_sweep( |
There was a problem hiding this comment.
nit: move this logic inside __attrs_post_init__
There was a problem hiding this comment.
I don't think this one could be moved to the attrs_post_init under ReadoutBenchmarkingParams? The reason is this is a specific validation if the experiment input is with sweep. However, the ReadoutBenchmarkingParams can be used by both sweep/non sweep functions.
| @@ -24,7 +24,10 @@ | |||
| import numpy as np | |||
|
|
|||
| from cirq import circuits, ops, work | |||
| from cirq.contrib.shuffle_circuits import run_shuffled_with_readout_benchmarking | |||
| from cirq.contrib.shuffle_circuits import run_shuffled_circuits_with_readout_benchmarking | |||
| from cirq.contrib.shuffle_circuits.shuffle_circuits_with_readout_benchmarking import ( | |||
There was a problem hiding this comment.
nit: lets use the Google convention here and import the modules
|
|
||
| def _determine_qubits_to_measure( | ||
| input_circuits: Sequence[circuits.Circuit], | ||
| qubits: Optional[Sequence[ops.Qid] | Sequence[Sequence[ops.Qid]]], |
There was a problem hiding this comment.
the type annotation for qubits is strange, why is a nested sequence?
There was a problem hiding this comment.
This is because each input circuit can have a corresponding qubit list to be measured. And since the function can accept a list of circuits, so it has a corresponding list[list[qid]], where each list[qid] matches a circuit.
| circuit_repetitions: int | list[int] | ||
| num_random_bitstrings: int = 100 | ||
| readout_repetitions: int = 1000 | ||
| qubits: Optional[Sequence[ops.Qid] | Sequence[Sequence[ops.Qid]]] = None |
There was a problem hiding this comment.
I know it was my suggestion, but thinking again about it I think qubits shouldn't be part of the ReadoutBenchmarkingParams class
| ) | ||
|
|
||
| for measurement in measurements: | ||
| assert isinstance(measurement, ResultDict) | ||
| if mode == "shuffled": |
There was a problem hiding this comment.
lets move these two logic parts into two functions _circuits_with_readout_benchmarking_errors_with_noise_shuffled and _circuits_with_readout_benchmarking_errors_with_noise_sweep
Requested by @NoureldinYosri, this pr is splited from #7358.
The pr adds a new function run_sweep_with_readout_benchmarking which runs the sweep circuits with readout error benchmarking (without shuffling).