Source code for pennylane.templates.embeddings.displacement

# Copyright 2018-2021 Xanadu Quantum Technologies Inc.

# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at

#     http://www.apache.org/licenses/LICENSE-2.0

# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
r"""
Contains the ``DisplacementEmbedding`` template.
"""
# pylint: disable-msg=too-many-branches,too-many-arguments,protected-access
import pennylane as qml
from pennylane.operation import AnyWires, Operation


class DisplacementEmbedding(Operation):
    r"""Encodes :math:`N` features into the displacement amplitudes :math:`r` or phases :math:`\phi` of :math:`M` modes,
    where :math:`N\leq M`.

    The mathematical definition of the displacement gate is given by the operator

    .. math::
            D(\alpha) = \exp(r (e^{i\phi}\ad -e^{-i\phi}\a)),

    where :math:`\a` and :math:`\ad` are the bosonic creation and annihilation operators.

    ``features`` has to be an array of at most ``len(wires)`` floats. If there are fewer entries in
    ``features`` than wires, the circuit does not apply the remaining displacement gates.

    Args:
        features (tensor_like): tensor of features
        wires (Any or Iterable[Any]): wires that the template acts on
        method (str): ``'phase'`` encodes the input into the phase of single-mode displacement, while
            ``'amplitude'`` uses the amplitude
        c (float): value of the phase of all displacement gates if ``execution='amplitude'``, or
            the amplitude of all displacement gates if ``execution='phase'``

    Raises:
        ValueError: if inputs do not have the correct format

    Example:

        Depending on the ``method`` argument, the feature vector will be encoded in the phase or the amplitude.
        The argument ``c`` will define the value of the other quantity.
        The default values are :math:`0.1` for ``c`` and ``'amplitude'`` for ``method``.

        .. code-block:: python

            dev = qml.device('default.gaussian', wires=3)

            @qml.qnode(dev)
            def circuit(feature_vector):
                qml.DisplacementEmbedding(features=feature_vector, wires=range(3))
                qml.QuadraticPhase(0.1, wires=1)
                return qml.expval(qml.NumberOperator(wires=1))

            X = [1, 2, 3]

        >>> print(circuit(X))
            4.1215690638748494

        And, the resulting circuit is:

        >>> print(qml.draw(circuit, show_matrices=False)(X))
        0: ─╭DisplacementEmbedding(M0)──────────┤
        1: ─├DisplacementEmbedding(M0)──P(0.10)─┤  <n>
        2: ─╰DisplacementEmbedding(M0)──────────┤

        Using different parameters:

        .. code-block:: python

            dev = qml.device('default.gaussian', wires=3)

            @qml.qnode(dev)
            def circuit(feature_vector):
                qml.DisplacementEmbedding(features=feature_vector, wires=range(3), method='phase', c=0.5)
                qml.QuadraticPhase(0.1, wires=1)
                return qml.expval(qml.NumberOperator(wires=1))

            X = [1, 2, 3]

        >>> print(circuit(X))
            0.23401288309122226

        And, the resulting circuit is:

        >>> print(qml.draw(circuit, show_matrices=False)(X))
        0: ─╭DisplacementEmbedding(M0)──────────┤
        1: ─├DisplacementEmbedding(M0)──P(0.10)─┤  <n>
        2: ─╰DisplacementEmbedding(M0)──────────┤
    """

    num_wires = AnyWires
    grad_method = None

    @classmethod
    def _unflatten(cls, data, metadata):
        new_op = cls.__new__(cls)
        Operation.__init__(new_op, *data, wires=metadata[0])
        return new_op

    def __init__(self, features, wires, method="amplitude", c=0.1, id=None):
        shape = qml.math.shape(features)
        constants = [c] * shape[0]
        constants = qml.math.convert_like(constants, features)

        if len(shape) != 1:
            raise ValueError(f"Features must be a one-dimensional tensor; got shape {shape}.")

        n_features = shape[0]
        if n_features != len(wires):
            raise ValueError(f"Features must be of length {len(wires)}; got length {n_features}.")

        if method == "amplitude":
            pars = qml.math.stack([features, constants], axis=1)

        elif method == "phase":
            pars = qml.math.stack([constants, features], axis=1)

        else:
            raise ValueError(f"did not recognize method {method}")

        super().__init__(pars, wires=wires, id=id)

    @property
    def num_params(self):
        return 1

    @staticmethod
    def compute_decomposition(pars, wires):  # pylint: disable=arguments-differ
        r"""Representation of the operator as a product of other operators.

        .. math:: O = O_1 O_2 \dots O_n.



        .. seealso:: :meth:`~.DisplacementEmbedding.decomposition`.

        Args:
            pars (tensor_like): parameters extracted from features and constant
            wires (Any or Iterable[Any]): wires that the template acts on

        Returns:
            list[.Operator]: decomposition of the operator

        **Example**

        >>> pars = torch.tensor([[1., 0.], [2., 0.]])
        >>> qml.DisplacementEmbedding.compute_decomposition(pars, wires=[0, 1])
        [Displacement(tensor(1.), tensor(0.), wires=[0]),
         Displacement(tensor(2.), tensor(0.), wires=[1])]
        """
        return [
            qml.Displacement(pars[i, 0], pars[i, 1], wires=wires[i : i + 1])
            for i in range(len(wires))
        ]

_modules/pennylane/templates/embeddings/displacement

 

Download Python script

 

Download Notebook

 

View on GitHub