iqm.error_reduction_tools.readout_characterization.readout_error_characterization.RECConfiguration

class iqm.error_reduction_tools.readout_characterization.readout_error_characterization.RECConfiguration(num_circuits=50, shots=10000, qubits=None, seed=None, symmetrize=True, equatorial_randomization=True)

Bases: object

Configuration for a readout error characterization (REC) experiment.

All parameters have sensible defaults so RECConfiguration() is valid and immediately usable.

Attributes

equatorial_randomization	When True, randomizes the phase angle of every X gate to average out coherent phase errors.
num_circuits	Number of calibration circuits to generate
qubits	Qubit names to characterize.
seed	Random seed for reproducible circuit generation.
shots	Total measurement shots distributed across all circuits.
symmetrize	When True, generates complementary preparation pairs so that each qubit is prepared in |0⟩ and |1⟩ equally often.

Methods

__eq__(other)	Return self==value.
__repr__()	Return repr(self).

Parameters:

• num_circuits (int)

• shots (int)

• qubits (list[str] | None)

• seed (int | None)

• symmetrize (bool)

• equatorial_randomization (bool)

num_circuits: int = 50

Number of calibration circuits to generate

shots: int = 10000

Total measurement shots distributed across all circuits. Each circuit receives shots // num_circuits shots.

qubits: list[str] | None = None

Qubit names to characterize.

Non-operational qubits (those without a calibrated measure gate) are always excluded: they cannot be characterized or have readout error mitigation applied.

• None (default): qubits are not pinned. When used standalone, this characterizes every operational qubit on the QPU; in the REMWorkflow it lets the workflow infer the qubits from the measured qubits of the input circuits.

• list[str]: characterize exactly these qubit names, minus any that are not operational.

seed: int | None = None

Random seed for reproducible circuit generation. None produces non-deterministic results.

symmetrize: bool = True

When True, generates complementary preparation pairs so that each qubit is prepared in |0⟩ and |1⟩ equally often.

equatorial_randomization: bool = True

When True, randomizes the phase angle of every X gate to average out coherent phase errors.