qml.QSVT¶
- class QSVT(UA, projectors)[source]¶
Bases:
pennylane.operation.Operation
Implements the quantum singular value transformation (QSVT) circuit.
Note
This template allows users to define hardware-compatible block encoding and projector-controlled phase shift circuits. For a QSVT implementation that is tailored for simulators see
qsvt()
.Given an
Operator
\(U\), which block encodes the matrix \(A\), and a list of projector-controlled phase shift operations \(\vec{\Pi}_\phi\), this template applies a circuit for the quantum singular value transformation as follows.When the number of projector-controlled phase shifts is even (\(d\) is odd), the QSVT circuit is defined as:
\[U_{QSVT} = \tilde{\Pi}_{\phi_1}U\left[\prod^{(d-1)/2}_{k=1}\Pi_{\phi_{2k}}U^\dagger \tilde{\Pi}_{\phi_{2k+1}}U\right]\Pi_{\phi_{d+1}}.\]And when the number of projector-controlled phase shifts is odd (\(d\) is even):
\[U_{QSVT} = \left[\prod^{d/2}_{k=1}\Pi_{\phi_{2k-1}}U^\dagger\tilde{\Pi}_{\phi_{2k}}U\right] \Pi_{\phi_{d+1}}.\]This circuit applies a polynomial transformation (\(Poly^{SV}\)) to the singular values of the block encoded matrix:
\[\begin{split}\begin{align} U_{QSVT}(A, \vec{\phi}) &= \begin{bmatrix} Poly^{SV}(A) & \cdot \\ \cdot & \cdot \end{bmatrix}. \end{align}\end{split}\]See also
- Parameters
UA (Operator) – the block encoding circuit, specified as an
Operator
, likeBlockEncode
projectors (Sequence[Operator]) – a list of projector-controlled phase shifts that implement the desired polynomial
- Raises
ValueError – if the input block encoding is not an operator
Example
To implement QSVT in a circuit, we can use the following method:
>>> dev = qml.device("default.qubit", wires=2) >>> A = np.array([[0.1]]) >>> block_encode = qml.BlockEncode(A, wires=[0, 1]) >>> shifts = [qml.PCPhase(i + 0.1, dim=1, wires=[0, 1]) for i in range(3)] >>> @qml.qnode(dev) >>> def example_circuit(): ... qml.QSVT(block_encode, shifts) ... return qml.expval(qml.Z(0))
The resulting circuit implements QSVT.
>>> print(qml.draw(example_circuit)()) 0: ─╭QSVT─┤ <Z> 1: ─╰QSVT─┤
To see the implementation details, we can expand the circuit:
>>> q_script = qml.tape.QuantumScript(ops=[qml.QSVT(block_encode, shifts)]) >>> print(q_script.expand().draw(decimals=2)) 0: ─╭∏_ϕ(0.10)─╭BlockEncode(M0)─╭∏_ϕ(1.10)─╭BlockEncode(M0)†─╭∏_ϕ(2.10)─┤ 1: ─╰∏_ϕ(0.10)─╰BlockEncode(M0)─╰∏_ϕ(1.10)─╰BlockEncode(M0)†─╰∏_ϕ(2.10)─┤
When working with this class directly, we can make use of any PennyLane operation to represent our block-encoding and our phase-shifts.
>>> dev = qml.device("default.qubit", wires=[0]) >>> block_encoding = qml.Hadamard(wires=0) # note H is a block encoding of 1/sqrt(2) >>> phase_shifts = [qml.RZ(-2 * theta, wires=0) for theta in (1.23, -0.5, 4)] # -2*theta to match convention >>> >>> @qml.qnode(dev) >>> def example_circuit(): ... qml.QSVT(block_encoding, phase_shifts) ... return qml.expval(qml.Z(0)) >>> >>> example_circuit() tensor(0.54030231, requires_grad=True)
Once again, we can visualize the circuit as follows:
>>> print(qml.draw(example_circuit)()) 0: ──QSVT─┤ <Z>
To see the implementation details, we can expand the circuit:
>>> q_script = qml.tape.QuantumScript(ops=[qml.QSVT(block_encoding, phase_shifts)]) >>> print(q_script.expand().draw(decimals=2)) 0: ──RZ(-2.46)──H──RZ(1.00)──H†──RZ(-8.00)─┤
Attributes
Arithmetic depth of the operator.
The basis of an operation, or for controlled gates, of the target operation.
Batch size of the operator if it is used with broadcasted parameters.
Control wires of the operator.
Flattened list of operator data in this QSVT operation.
Gradient computation method.
Gradient recipe for the parameter-shift method.
Integer hash that uniquely represents the operator.
Dictionary of non-trainable variables that this operation depends on.
Custom string to label a specific operator instance.
This property determines if an operator is hermitian.
String for the name of the operator.
Number of dimensions per trainable parameter of the operator.
Number of trainable parameters that the operator depends on.
Number of wires that the operator acts on.
Returns the frequencies for each operator parameter with respect to an expectation value of the form \(\langle \psi | U(\mathbf{p})^\dagger \hat{O} U(\mathbf{p})|\psi\rangle\).
Trainable parameters that the operator depends on.
A
PauliSentence
representation of the Operator, orNone
if it doesn't have one.Wires that the operator acts on.
- arithmetic_depth¶
Arithmetic depth of the operator.
- basis¶
The basis of an operation, or for controlled gates, of the target operation. If not
None
, should take a value of"X"
,"Y"
, or"Z"
.For example,
X
andCNOT
havebasis = "X"
, whereasControlledPhaseShift
andRZ
havebasis = "Z"
.- Type
str or None
- batch_size¶
Batch size of the operator if it is used with broadcasted parameters.
The
batch_size
is determined based onndim_params
and the provided parameters for the operator. If (some of) the latter have an additional dimension, and this dimension has the same size for all parameters, its size is the batch size of the operator. If no parameter has an additional dimension, the batch size isNone
.- Returns
Size of the parameter broadcasting dimension if present, else
None
.- Return type
int or None
- control_wires¶
Control wires of the operator.
For operations that are not controlled, this is an empty
Wires
object of length0
.- Returns
The control wires of the operation.
- Return type
- data¶
Flattened list of operator data in this QSVT operation.
This ensures that the backend of a
QuantumScript
which contains aQSVT
operation can be inferred with respect to the types of theQSVT
block encoding and projector-controlled phase shift data.
- grad_method = None¶
Gradient computation method.
- grad_recipe = None¶
Gradient recipe for the parameter-shift method.
This is a tuple with one nested list per operation parameter. For parameter \(\phi_k\), the nested list contains elements of the form \([c_i, a_i, s_i]\) where \(i\) is the index of the term, resulting in a gradient recipe of
\[\frac{\partial}{\partial\phi_k}f = \sum_{i} c_i f(a_i \phi_k + s_i).\]If
None
, the default gradient recipe containing the two terms \([c_0, a_0, s_0]=[1/2, 1, \pi/2]\) and \([c_1, a_1, s_1]=[-1/2, 1, -\pi/2]\) is assumed for every parameter.- Type
tuple(Union(list[list[float]], None)) or None
- has_adjoint = False¶
- has_decomposition = True¶
- has_diagonalizing_gates = False¶
- has_generator = False¶
- has_matrix = True¶
- has_sparse_matrix = False¶
- hash¶
Integer hash that uniquely represents the operator.
- Type
int
- hyperparameters¶
Dictionary of non-trainable variables that this operation depends on.
- Type
dict
- id¶
Custom string to label a specific operator instance.
- is_hermitian¶
This property determines if an operator is hermitian.
- name¶
String for the name of the operator.
- ndim_params¶
Number of dimensions per trainable parameter of the operator.
By default, this property returns the numbers of dimensions of the parameters used for the operator creation. If the parameter sizes for an operator subclass are fixed, this property can be overwritten to return the fixed value.
- Returns
Number of dimensions for each trainable parameter.
- Return type
tuple
- num_params¶
Number of trainable parameters that the operator depends on.
By default, this property returns as many parameters as were used for the operator creation. If the number of parameters for an operator subclass is fixed, this property can be overwritten to return the fixed value.
- Returns
number of parameters
- Return type
int
- num_wires = -1¶
Number of wires that the operator acts on.
- Type
int
- parameter_frequencies¶
Returns the frequencies for each operator parameter with respect to an expectation value of the form \(\langle \psi | U(\mathbf{p})^\dagger \hat{O} U(\mathbf{p})|\psi\rangle\).
These frequencies encode the behaviour of the operator \(U(\mathbf{p})\) on the value of the expectation value as the parameters are modified. For more details, please see the
pennylane.fourier
module.- Returns
Tuple of frequencies for each parameter. Note that only non-negative frequency values are returned.
- Return type
list[tuple[int or float]]
Example
>>> op = qml.CRot(0.4, 0.1, 0.3, wires=[0, 1]) >>> op.parameter_frequencies [(0.5, 1), (0.5, 1), (0.5, 1)]
For operators that define a generator, the parameter frequencies are directly related to the eigenvalues of the generator:
>>> op = qml.ControlledPhaseShift(0.1, wires=[0, 1]) >>> op.parameter_frequencies [(1,)] >>> gen = qml.generator(op, format="observable") >>> gen_eigvals = qml.eigvals(gen) >>> qml.gradients.eigvals_to_frequencies(tuple(gen_eigvals)) (1.0,)
For more details on this relationship, see
eigvals_to_frequencies()
.
- parameters¶
Trainable parameters that the operator depends on.
- pauli_rep¶
A
PauliSentence
representation of the Operator, orNone
if it doesn’t have one.
Methods
adjoint
()Create an operation that is the adjoint of this one.
compute_decomposition
(*_data, UA, ...)Representation of the operator as a product of other operators.
compute_diagonalizing_gates
(*params, wires, ...)Sequence of gates that diagonalize the operator in the computational basis (static method).
compute_eigvals
(*params, **hyperparams)Eigenvalues of the operator in the computational basis (static method).
compute_matrix
(*args, **kwargs)Representation of the operator as a canonical matrix in the computational basis (static method).
compute_sparse_matrix
(*params, **hyperparams)Representation of the operator as a sparse matrix in the computational basis (static method).
Representation of the operator as a product of other operators.
Sequence of gates that diagonalize the operator in the computational basis.
eigvals
()Eigenvalues of the operator in the computational basis.
Generator of an operator that is in single-parameter-form.
label
([decimals, base_label, cache])A customizable string representation of the operator.
map_wires
(wire_map)Returns a copy of the current operator with its wires changed according to the given wire map.
matrix
([wire_order])Representation of the operator as a matrix in the computational basis.
pow
(z)A list of new operators equal to this one raised to the given power.
queue
([context])Append the operator to the Operator queue.
simplify
()Reduce the depth of nested operators to the minimum.
The parameters required to implement a single-qubit gate as an equivalent
Rot
gate, up to a global phase.sparse_matrix
([wire_order])Representation of the operator as a sparse matrix in the computational basis.
terms
()Representation of the operator as a linear combination of other operators.
- adjoint()¶
Create an operation that is the adjoint of this one.
Adjointed operations are the conjugated and transposed version of the original operation. Adjointed ops are equivalent to the inverted operation for unitary gates.
- Returns
The adjointed operation.
- static compute_decomposition(*_data, UA, projectors, **_kwargs)[source]¶
Representation of the operator as a product of other operators.
The
QSVT
is decomposed into alternating block encoding and projector-controlled phase shift operators. This is defined by the following equations, where \(U\) is the block encoding operation and both \(\Pi_\phi\) and \(\tilde{\Pi}_\phi\) are projector-controlled phase shifts with angle \(\phi\).When the number of projector-controlled phase shifts is even (\(d\) is odd), the QSVT circuit is defined as:
\[U_{QSVT} = \Pi_{\phi_1}U\left[\prod^{(d-1)/2}_{k=1}\Pi_{\phi_{2k}}U^\dagger \tilde{\Pi}_{\phi_{2k+1}}U\right]\Pi_{\phi_{d+1}}.\]And when the number of projector-controlled phase shifts is odd (\(d\) is even):
\[U_{QSVT} = \left[\prod^{d/2}_{k=1}\Pi_{\phi_{2k-1}}U^\dagger\tilde{\Pi}_{\phi_{2k}}U\right] \Pi_{\phi_{d+1}}.\]See also
- static compute_diagonalizing_gates(*params, wires, **hyperparams)¶
Sequence of gates that diagonalize the operator in the computational basis (static method).
Given the eigendecomposition \(O = U \Sigma U^{\dagger}\) where \(\Sigma\) is a diagonal matrix containing the eigenvalues, the sequence of diagonalizing gates implements the unitary \(U^{\dagger}\).
The diagonalizing gates rotate the state into the eigenbasis of the operator.
See also
- Parameters
params (list) – trainable parameters of the operator, as stored in the
parameters
attributewires (Iterable[Any], Wires) – wires that the operator acts on
hyperparams (dict) – non-trainable hyperparameters of the operator, as stored in the
hyperparameters
attribute
- Returns
list of diagonalizing gates
- Return type
list[Operator]
- static compute_eigvals(*params, **hyperparams)¶
Eigenvalues of the operator in the computational basis (static method).
If
diagonalizing_gates
are specified and implement a unitary \(U^{\dagger}\), the operator can be reconstructed as\[O = U \Sigma U^{\dagger},\]where \(\Sigma\) is the diagonal matrix containing the eigenvalues.
Otherwise, no particular order for the eigenvalues is guaranteed.
See also
- Parameters
*params (list) – trainable parameters of the operator, as stored in the
parameters
attribute**hyperparams (dict) – non-trainable hyperparameters of the operator, as stored in the
hyperparameters
attribute
- Returns
eigenvalues
- Return type
tensor_like
- static compute_matrix(*args, **kwargs)[source]¶
Representation of the operator as a canonical matrix in the computational basis (static method).
The canonical matrix is the textbook matrix representation that does not consider wires. Implicitly, this assumes that the wires of the operator correspond to the global wire order.
- Parameters
*params (list) – trainable parameters of the operator, as stored in the
parameters
attribute**hyperparams (dict) – non-trainable hyperparameters of the operator, as stored in the
hyperparameters
attribute
- Returns
matrix representation
- Return type
tensor_like
- static compute_sparse_matrix(*params, **hyperparams)¶
Representation of the operator as a sparse matrix in the computational basis (static method).
The canonical matrix is the textbook matrix representation that does not consider wires. Implicitly, this assumes that the wires of the operator correspond to the global wire order.
See also
- Parameters
*params (list) – trainable parameters of the operator, as stored in the
parameters
attribute**hyperparams (dict) – non-trainable hyperparameters of the operator, as stored in the
hyperparameters
attribute
- Returns
sparse matrix representation
- Return type
scipy.sparse._csr.csr_matrix
- decomposition()¶
Representation of the operator as a product of other operators.
\[O = O_1 O_2 \dots O_n\]A
DecompositionUndefinedError
is raised if no representation by decomposition is defined.See also
- Returns
decomposition of the operator
- Return type
list[Operator]
- diagonalizing_gates()¶
Sequence of gates that diagonalize the operator in the computational basis.
Given the eigendecomposition \(O = U \Sigma U^{\dagger}\) where \(\Sigma\) is a diagonal matrix containing the eigenvalues, the sequence of diagonalizing gates implements the unitary \(U^{\dagger}\).
The diagonalizing gates rotate the state into the eigenbasis of the operator.
A
DiagGatesUndefinedError
is raised if no representation by decomposition is defined.See also
- Returns
a list of operators
- Return type
list[Operator] or None
- eigvals()¶
Eigenvalues of the operator in the computational basis.
If
diagonalizing_gates
are specified and implement a unitary \(U^{\dagger}\), the operator can be reconstructed as\[O = U \Sigma U^{\dagger},\]where \(\Sigma\) is the diagonal matrix containing the eigenvalues.
Otherwise, no particular order for the eigenvalues is guaranteed.
Note
When eigenvalues are not explicitly defined, they are computed automatically from the matrix representation. Currently, this computation is not differentiable.
A
EigvalsUndefinedError
is raised if the eigenvalues have not been defined and cannot be inferred from the matrix representation.See also
- Returns
eigenvalues
- Return type
tensor_like
- generator()¶
Generator of an operator that is in single-parameter-form.
For example, for operator
\[U(\phi) = e^{i\phi (0.5 Y + Z\otimes X)}\]we get the generator
>>> U.generator() 0.5 * Y(0) + Z(0) @ X(1)
The generator may also be provided in the form of a dense or sparse Hamiltonian (using
Hamiltonian
andSparseHamiltonian
respectively).The default value to return is
None
, indicating that the operation has no defined generator.
- label(decimals=None, base_label=None, cache=None)[source]¶
A customizable string representation of the operator.
- Parameters
decimals=None (int) – If
None
, no parameters are included. Else, specifies how to round the parameters.base_label=None (str) – overwrite the non-parameter component of the label
cache=None (dict) – dictionary that carries information between label calls in the same drawing
- Returns
label to use in drawings
- Return type
str
Example:
>>> op = qml.RX(1.23456, wires=0) >>> op.label() "RX" >>> op.label(base_label="my_label") "my_label" >>> op = qml.RX(1.23456, wires=0, id="test_data") >>> op.label() "RX("test_data")" >>> op.label(decimals=2) "RX\n(1.23,"test_data")" >>> op.label(base_label="my_label") "my_label("test_data")" >>> op.label(decimals=2, base_label="my_label") "my_label\n(1.23,"test_data")"
If the operation has a matrix-valued parameter and a cache dictionary is provided, unique matrices will be cached in the
'matrices'
key list. The label will contain the index of the matrix in the'matrices'
list.>>> op2 = qml.QubitUnitary(np.eye(2), wires=0) >>> cache = {'matrices': []} >>> op2.label(cache=cache) 'U(M0)' >>> cache['matrices'] [tensor([[1., 0.], [0., 1.]], requires_grad=True)] >>> op3 = qml.QubitUnitary(np.eye(4), wires=(0,1)) >>> op3.label(cache=cache) 'U(M1)' >>> cache['matrices'] [tensor([[1., 0.], [0., 1.]], requires_grad=True), tensor([[1., 0., 0., 0.], [0., 1., 0., 0.], [0., 0., 1., 0.], [0., 0., 0., 1.]], requires_grad=True)]
- map_wires(wire_map)[source]¶
Returns a copy of the current operator with its wires changed according to the given wire map.
- Parameters
wire_map (dict) – dictionary containing the old wires as keys and the new wires as values
- Returns
new operator
- Return type
- matrix(wire_order=None)¶
Representation of the operator as a matrix in the computational basis.
If
wire_order
is provided, the numerical representation considers the position of the operator’s wires in the global wire order. Otherwise, the wire order defaults to the operator’s wires.If the matrix depends on trainable parameters, the result will be cast in the same autodifferentiation framework as the parameters.
A
MatrixUndefinedError
is raised if the matrix representation has not been defined.See also
- Parameters
wire_order (Iterable) – global wire order, must contain all wire labels from the operator’s wires
- Returns
matrix representation
- Return type
tensor_like
- pow(z)¶
A list of new operators equal to this one raised to the given power.
- Parameters
z (float) – exponent for the operator
- Returns
list[
Operator
]
- queue(context=<class 'pennylane.queuing.QueuingManager'>)[source]¶
Append the operator to the Operator queue.
- simplify()¶
Reduce the depth of nested operators to the minimum.
- Returns
simplified operator
- Return type
- single_qubit_rot_angles()¶
The parameters required to implement a single-qubit gate as an equivalent
Rot
gate, up to a global phase.- Returns
A list of values \([\phi, \theta, \omega]\) such that \(RZ(\omega) RY(\theta) RZ(\phi)\) is equivalent to the original operation.
- Return type
tuple[float, float, float]
- sparse_matrix(wire_order=None)¶
Representation of the operator as a sparse matrix in the computational basis.
If
wire_order
is provided, the numerical representation considers the position of the operator’s wires in the global wire order. Otherwise, the wire order defaults to the operator’s wires.A
SparseMatrixUndefinedError
is raised if the sparse matrix representation has not been defined.See also
- Parameters
wire_order (Iterable) – global wire order, must contain all wire labels from the operator’s wires
- Returns
sparse matrix representation
- Return type
scipy.sparse._csr.csr_matrix
- terms()¶
Representation of the operator as a linear combination of other operators.
\[O = \sum_i c_i O_i\]A
TermsUndefinedError
is raised if no representation by terms is defined.- Returns
list of coefficients \(c_i\) and list of operations \(O_i\)
- Return type
tuple[list[tensor_like or float], list[Operation]]