AngleEmbedding#

class pyqit.core.embeddings.AngleEmbedding(n_qubits: int, rotation: str = 'X')[source]#

Bases: BaseEmbedding

One rotation per qubit, PennyLane’s AngleEmbedding.

Takes one feature per wire. The DataModule zero-pads narrower input to n_qubits columns and multiplies by pi, which maps features normalized to [0, 1] onto [0, pi]. Input wider than n_qubits raises.

Parameters:

  • n_qubits (int)

  • rotation ({"X", "Y", "Z"}, default "X") – Rotation gate the features drive.

Examples

>>> from pyqit.core import AngleEmbedding
>>> from pyqit.models import VQCClassifier
>>> model = VQCClassifier(n_qubits=4, encoder=AngleEmbedding)
forward(inputs)[source]#

Apply one rotation gate per qubit. Expects inputs scaled to [0, pi].

classmethod get_test_params()[source]#

List constructor kwargs used to parametrize this class in the test suite.