API reference

Classification

QuanticClassifierBase([quantum, ...])	Quantum classifier
QuanticSVM([gamma, C, max_iter, pegasos, ...])	Quantum-enhanced SVM classifier
QuanticVQC([optimizer, gen_var_form, ...])	Variational quantum classifier
QuanticMDM([metric, quantum, ...])	Quantum-enhanced MDM classifier
QuanticNCH([quantum, q_account_token, ...])	A Quantum wrapper around the NCH algorithm.
NearestConvexHull([n_jobs, ...])	Classification by Nearest Convex Hull (NCH).
ContinuousQIOCEClassifier([n_reps, ...])	QAOA classifier with batch training using angle encoding.
QuantumStateDiscriminator([n_jobs])	Quantum state classifier using the Pretty Good Measurement (PGM).
CpMDM([optimizer])	Quantum-enhanced MDM classifier

Autoencoders

BasicQnnAutoencoder([num_latent, num_trash, ...])

Quantum denoising

Pipelines

BasePipeline(code)	Base class for quantum classifiers with Riemannian pipeline.
QuantumClassifierWithDefaultRiemannianPipeline([...])	Default pipeline with Riemannian geometry and a quantum classifier.
QuantumMDMWithRiemannianPipeline([metric, ...])	MDM with Riemannian pipeline adapted for cpm metrics.
QuantumMDMVotingClassifier([quantum, ...])	Voting classifier with two QuantumMDMWithRiemannianPipeline
FeaturesUnionClassifier([transformers, ...])	An alias for FeatureUnion + Classifier

Ensemble

JudgeClassifier(judge, clfs)

Judge classifier

Optimization

Optimization contains quantum and classical optimizers and supporting mathematical utilities.

Distance

qdistance_logeuclid_to_convex_hull(A, B[, ...])	Log-Euclidean distance to a convex hull of SPD matrices.
weights_logeuclid_to_convex_hull(A, B[, ...])	Weights for Log-Euclidean distance to a convex hull of SPD matrices.

Mean

qmean_euclid(X[, sample_weight, optimizer])	Euclidean mean with Constraint Programming Model.
qmean_logeuclid(X[, sample_weight, optimizer])	Log-Euclidean mean with Constraint Programming Model.

Docplex

square_cont_mat_var(prob, channels[, name])	Docplex square continuous matrix
square_int_mat_var(prob, channels[, ...])	Docplex square integer matrix
square_bin_mat_var(prob, channels[, name])	Docplex square binary matrix
pyQiskitOptimizer()	Wrapper for Qiskit optimizer.
ClassicalOptimizer([optimizer])	Wrapper for the classical Cobyla optimizer.
NaiveQAOAOptimizer([upper_bound, ...])	Wrapper for the quantum optimizer QAOA.
QAOACVAngleOptimizer([create_mixer, n_reps, ...])	QAOA with continuous variables encoded in state vector angles.
QAOACVOptimizer([create_mixer, n_reps, ...])	QAOA with continuous variables.

Optimizers

anderson_optimizer.AndersonAccelerationOptimizer([...])	Anderson acceleration optimizer for variational quantum circuits.
riemannian_adam.RiemannianAdamOptimizer([...])	Adam optimizer with manifold-aware retraction for VQC parameters, inspired by [Rf2ef09536214-1].

Utils functions

Utils functions are low level functions for the classification and pipelines module.

Utils

is_qfunction(string)	Indicates if the function is a mean or a distance introduced in this library.
get_docplex_optimizer_from_params_bag(...)	Factory function to create optimizer based on parameters.

Hyper-parameters generation

gen_x_feature_map([reps])	Return a callable that generates a XFeatureMap.
gen_z_feature_map([reps])	Return a callable that generates a ZFeatureMap.
gen_zz_feature_map([reps, entanglement])	Return a callable that generates a ZZFeatureMap.
gen_two_local([reps, rotation_blocks, ...])	Return a callable that generate a TwoLocal circuit.
get_spsa([max_trials, c])	Return an instance of SPSA.
get_spsa_parameters(spsa)	Return the default values of the calibrate method of an SPSA instance.
create_mixer_rotational_X_gates(angle)	Return the default mixing operator with QAOA.
create_mixer_rotational_XY_gates(angle)	Return the XY mixer.
create_mixer_rotational_XZ_gates(angle)	Return a mixing operator with XZ gates.
create_mixer_qiskit_default(_angle)	Return the default mixing operator with QAOA.
create_mixer_with_circular_entanglement(angle)	Return a mixer with rotation gates and circular entanglement.
create_mixer_identity()	Return an identity mixer (no operation).

Preprocessing

NdRobustScaler()	Apply one robust scaler by feature.
Vectorizer()	Vectorization.
Devectorizer(n_features, n_samples)	Transform vector to matrices

Filtering

NoDimRed()	No dimensional reduction.
NaiveDimRed([is_even])	Naive dimensional reduction.
ChannelSelection(n_channels[, cov_est])	Select channel in epochs.

Math

union_of_diff(*arrays)	Return the positions for which at least one of the array as a different value than the others.
to_xyz(X)	Plot histogram of bi-class predictions.
is_pauli_identity(operator)	Check if operator is pauli identity

Datasets

get_mne_sample([n_trials, include_auditory])	Return sample data from the MNE dataset [R704a40a78b24-1].
generate_linearly_separable_dataset([n_samples])	Return a linearly separable dataset.
generate_qiskit_dataset([n_samples])	Return a Qiskit ad-hoc dataset.
get_feature_dimension(dataset)	Return the feature dimension of a dataset.
MockDataset(dataset_gen, n_subjects)	A dataset with mock data.

Quantum Provider

SymbFidelityStatevectorKernel(feature_map, ...)	Symbolic Statevector kernel
get_provider()	Return an IBM quantum provider.
get_device(provider, min_qubits)	Returns all real remote quantum backends.
get_simulator()	Return a quantum simulator.
get_quantum_kernel(feature_map, ...[, n_jobs])	Get a quantum kernel

Transfer Learning

Adapter(preprocessing, estimator)	Meta-estimator bridging MOABB's evaluation interface with pyriemann TL pipelines.
TLCrossSubjectEvaluation(paradigm[, ...])	CrossSubjectEvaluation with group and target-domain forwarding.

Visualization

Helpers to visualize distances, manifold and even “artistic” representation.

Art

weights_spiral(axe, vqc, X, y[, n_trainings])

Artistic representation of vqc training.

Manifold

plot_cvx_hull(X, ax)	Plot the convex hull of a set of points.
plot_manifold(X, y[, plot_hull])	Plot spd matrices in 3d (cartesian coordinate system).