Optimizers

Divi provides multiple optimization strategies for quantum algorithm parameter tuning, from classical gradient-based methods to quantum-inspired approaches.

divi.qprog.optimizers Package

Classes

FubiniStudyMetricEstimator()	Block-diagonal Fubini–Study metric (quantum geometric tensor).
GridSearchOptimizer([param_grid, ...])	Exhaustive grid search optimizer.
MetricEstimator()	Strategy that produces natural-gradient evaluators for a program.
MonteCarloOptimizer([population_size, ...])	Monte Carlo-based parameter search optimizer.
MonteCarloState(**data)	Pydantic model for Monte Carlo optimizer state.
Optimizer()	Abstract base class for all optimizers.
PullbackMetricEstimator()	Hamiltonian-aware pullback metric.
PymooMethod()	Supported optimization methods from the pymoo library.
PymooOptimizer(method[, population_size])	Optimizer wrapper for pymoo optimization algorithms and CMA-ES.
PymooState(**data)	Pydantic model for Pymoo optimizer state.
QNGOptimizer([step_size, regularization, ...])	Quantum Natural Gradient optimizer.
QNSPSAOptimizer([learning_rate, c, alpha, ...])	Quantum Natural SPSA (Gacon et al.).
QUIVEROptimizer([learning_rate, epsilon, ...])	Adaptive directional (forward) gradients — QUIVER (arXiv 2606.09734).
SPSAOptimizer([learning_rate, c, alpha, ...])	Simultaneous Perturbation Stochastic Approximation (Spall).
ScipyMethod()	Supported optimization methods from scipy.optimize.
ScipyOptimizer(method)	Optimizer wrapper for scipy.optimize methods.
StochasticFidelityMetricEstimator()	Stochastic Fubini–Study metric via state-overlap fidelities (QN-SPSA).