qlauncher.routines.qiskit.algorithms.qiskit_native

Algorithms for Qiskit routines

Summary

Classes:

FALQON	Algorithm class with FALQON.
QAOA	Algorithm class with QAOA.
QiskitOptimizationAlgorithm	Abstract class for Qiskit optimization algorithms
VQE	Variational Quantum EigenSolver - qiskit-algorithm implementation wrapper.

Functions:

commutator	Commutator
cvar_cost	CVar cost function to be used instead of mean for qaoa training
int_to_bitstring

Reference

class qlauncher.routines.qiskit.algorithms.qiskit_native.QiskitOptimizationAlgorithm(**alg_kwargs)

Bases: Algorithm

Abstract class for Qiskit optimization algorithms

make_tag(problem: Problem, backend: QiskitBackend) → str

get_processing_times(tag: str, primitive: BasePrimitive) → None | tuple[list, list, int]

qlauncher.routines.qiskit.algorithms.qiskit_native.commutator(op_a: SparsePauliOp, op_b: SparsePauliOp) → SparsePauliOp

Commutator

qlauncher.routines.qiskit.algorithms.qiskit_native.int_to_bitstring(number: int, total_bits: int)

qlauncher.routines.qiskit.algorithms.qiskit_native.cvar_cost(probs_values: Iterable[tuple[float, float]], alpha: float)

CVar cost function to be used instead of mean for qaoa training

class qlauncher.routines.qiskit.algorithms.qiskit_native.QAOA(p: int = 1, optimization_method: Literal['COBYLA'] = 'COBYLA', max_evaluations: int = 100, training_aggregation_method: Literal['mean', 'cvar'] = 'mean', cvar_alpha: float = 1, alternating_ansatz: bool = False, aux=None, **alg_kwargs)

Bases: QiskitOptimizationAlgorithm

Algorithm class with QAOA.

Parameters:

• p (int) – The number of QAOA steps. Defaults to 1.

• optimizer (Optimizer | None) – Optimizer used during algorithm runtime. If set to None turns into COBYLA. Defaults to None,

• alternating_ansatz (bool) – Whether to use an alternating ansatz. Defaults to False. If True, it’s recommended to provide a mixer_h to alg_kwargs.

• aux – Auxiliary input for the QAOA algorithm.

• **alg_kwargs – Additional keyword arguments for the base class.

name

The name of the algorithm.

Type:

str

aux

Auxiliary input for the QAOA algorithm.

p

The number of QAOA steps.

Type:

int

optimizer

Optimizer used during algorithm runtime.

Type:

Optimizer

alternating_ansatz

Whether to use an alternating ansatz.

Type:

bool

parameters

List of parameters for the algorithm.

Type:

list

mixer_h

The mixer Hamiltonian.

Type:

SparsePauliOp | None

property setup: dict

run(problem: Problem, backend: Backend, formatter: Callable) → Result

Runs the QAOA algorithm

construct_result(result: dict) → Result

class qlauncher.routines.qiskit.algorithms.qiskit_native.FALQON(driver_h: SparsePauliOp | None = None, delta_t: float = 0.03, beta_0: float = 0.0, max_reps: int = 20)

Bases: QiskitOptimizationAlgorithm

Algorithm class with FALQON.

Parameters:

• driver_h (Operator | None) – The driver Hamiltonian for the problem.

• delta_t (float) – The time step for the evolution operators.

• beta_0 (float) – The initial value of beta.

• n (int) – The number of iterations to run the algorithm.

• **alg_kwargs – Additional keyword arguments for the base class.

driver_h

The driver Hamiltonian for the problem.

Type:

Operator | None

delta_t

The time step for the evolution operators.

Type:

float

beta_0

The initial value of beta.

Type:

float

n

The number of iterations to run the algorithm.

Type:

int

cost_h

The cost Hamiltonian for the problem.

Type:

Operator | None

n_qubits

The number of qubits in the problem.

Type:

int

parameters

The list of algorithm parameters.

Type:

list[str]

property setup: dict

run(problem: Problem, backend: QiskitBackend, formatter: Callable) → Result

Runs the FALQON algorithm

class qlauncher.routines.qiskit.algorithms.qiskit_native.VQE(optimizer: Optimizer | None = None, ansatz: QuantumCircuit | None = None, with_numpy: bool = False)

Bases: QiskitOptimizationAlgorithm

Variational Quantum EigenSolver - qiskit-algorithm implementation wrapper.

Parameters:

• optimizer (optimizers.Optimizer | None, optional) – Optimizer for VQE. Defaults to None.

• ansatz (QuantumCircuit | None, optional) – VQE’s ansatz. Defaults to None.

• with_numpy (bool, optional) – Ignores ansatz parameter and backend, and changes solver to Numpy based. Defaults to False.

property ansatz: QuantumCircuit

run(problem: Problem, backend: Backend, formatter: Callable[[...], Any]) → Result

Runs the algorithm on a specific problem using a backend.

Parameters:

• problem (Problem) – The problem to be solved.

• backend (Backend) – The backend to be used for execution.

construct_result(result: EigenstateResult) → Result