uniqc.algorithms.core.training.hybrid_model module

Hybrid Classical-Quantum model with TorchQuantum backend.

Combines classical neural network layers with a quantum circuit layer for hybrid quantum-classical machine learning.

class uniqc.algorithms.core.training.hybrid_model.HybridQCLModel(n_features=2, n_qubits=4, quantum_depth=2, classical_hidden=32)

Bases: Module

Hybrid Classical-Quantum model.

Architecture: ClassicalEncoder → QuantumLayer → ClassicalDecoder

Parameters:

• n_features – Input feature dimension.

• n_qubits – Number of quantum circuit qubits.

• quantum_depth – HEA depth for quantum layer.

• classical_hidden – Hidden layer size for classical nets.

forward(x)

Forward pass through hybrid model.

Parameters:

x – Input tensor of shape (batch_size, n_features).

Returns:

Output tensor of shape (batch_size, 1).