pyriemann_qiskit.autoencoders.BasicQnnAutoencoder

class pyriemann_qiskit.autoencoders.BasicQnnAutoencoder(num_latent=3, num_trash=2, opt=<qiskit_algorithms.optimizers.spsa.SPSA object>, callback=None)[source]

Quantum denoising

This class implements a quantum auto encoder. The implementation was adapted from [1], to be compatible with scikit-learn.

Parameters:

  • num_latent (int, default=3) – The number of qubits in the latent space.

  • num_trash (int, default=2) – The number of qubits in the trash space.

  • opt (Optimizer, default=SPSA(maxiter=100, blocking=True)) – The classical optimizer to use.

  • callback (Callable[int, double], default=None) – An additional callback for the optimizer. The first parameter is the number of cost evaluation call. The second parameter is the cost.

Notes

Added in version 0.3.0.

costs_

The values of the cost function.

Type:

list

fidelities_

fidelities (one fidelity for each sample).

Type:

list, shape (n_samples,)

References

[2]

A. Mostafa et al., 2024 ‘Quantum Denoising in the Realm of Brain-Computer Interfaces: A Preliminary Study’, https://hal.science/hal-04501908

__init__(num_latent=3, num_trash=2, opt=<qiskit_algorithms.optimizers.spsa.SPSA object>, callback=None)[source]

Examples using pyriemann_qiskit.autoencoders.BasicQnnAutoencoder

Quantum autoencoder for signal denoising

Quantum autoencoder for signal denoising