qml.CircuitGraph

class CircuitGraph(ops, obs, wires, par_info=None, trainable_params=None)

Bases: object

Represents a quantum circuit as a directed acyclic graph.

In this representation the Operator instances are the nodes of the graph, and each directed edge represent a subsystem (or a group of subsystems) on which the two Operators act subsequently. This representation can describe the causal relationships between arbitrary quantum channels and measurements, not just unitary gates.

Parameters:

• ops (Iterable[.Operator]) – quantum operators constituting the circuit, in temporal order

• obs (list[Union[MeasurementProcess, Operator]]) – terminal measurements, in temporal order

• wires (.Wires) – The addressable wire registers of the device that will be executing this graph

• par_info (Optional[list[dict]]) – Parameter information. For each index, the entry is a dictionary containing an operation

• parameters. (and an index into that operation's)

• trainable_params (Optional[set[int]]) – A set containing the indices of parameters that support differentiability. The indices provided match the order of appearance in the quantum circuit.

Attributes

graph	The graph representation of the quantum circuit.
hash	Creating a hash for the circuit graph based on the string generated by serialize.
max_simultaneous_measurements	Returns the maximum number of measurements on any wire in the circuit graph.
observables	Observables in the circuit.
observables_in_order	Observables in the circuit, in a fixed topological order.
operations	Operations in the circuit.
operations_in_order	Operations in the circuit, in a fixed topological order.
parametrized_layers	Identify the parametrized layer structure of the circuit.

graph

The graph representation of the quantum circuit.

The graph has nodes representing indices into the queue, and directed edges pointing from nodes to their immediate dependents/successors.

Returns:

the directed acyclic graph representing the quantum circuit

Return type:

rustworkx.PyDiGraph

hash

Creating a hash for the circuit graph based on the string generated by serialize.

Returns:

the hash of the serialized quantum circuit graph

Return type:

int

max_simultaneous_measurements

Returns the maximum number of measurements on any wire in the circuit graph.

This method counts the number of measurements for each wire and returns the maximum.

Examples

>>> dev = qml.device('default.qubit', wires=3)
>>> def circuit_measure_max_once():
...     return qml.expval(qml.X(0))
>>> qnode = qml.QNode(circuit_measure_max_once, dev)
>>> tape = qml.workflow.construct_tape(qnode)()
>>> tape.graph.max_simultaneous_measurements
1
>>> def circuit_measure_max_twice():
...     return qml.expval(qml.X(0)), qml.probs(wires=0)
>>> qnode = qml.QNode(circuit_measure_max_twice, dev)
>>> tape = qml.workflow.construct_tape(qnode)()
>>> tape.graph.max_simultaneous_measurements
2

Returns:

the maximum number of measurements

Return type:

int

observables

Observables in the circuit.

observables_in_order

Observables in the circuit, in a fixed topological order.

The topological order used by this method is guaranteed to be the same as the order in which the measured observables are returned by the quantum function. Currently the topological order is determined by the queue index.

Returns:

observables

Return type:

list[Union[MeasurementProcess, Operator]]

operations

Operations in the circuit.

operations_in_order

Operations in the circuit, in a fixed topological order.

Currently the topological order is determined by the queue index.

The complement of QNode.observables(). Together they return every Operator instance in the circuit.

Returns:

operations

Return type:

list[Operation]

parametrized_layers

Identify the parametrized layer structure of the circuit.

Returns:

layers of the circuit

Return type:

list[Layer]

Methods

ancestors(ops[, sort])	Ancestors of a given set of operators.
ancestors_in_order(ops)	Operator ancestors in a topological order.
ancestors_of_indexes(indexes[, sort])	Ancestors of a given set of operators.
descendants(ops[, sort])	Descendants of a given set of operators.
descendants_in_order(ops)	Operator descendants in a topological order.
descendants_of_indexes(indexes[, sort])	Descendants of a given set of operators.
get_depth()	Depth of the quantum circuit (longest path in the DAG).
has_path(a, b)	Checks if a path exists between the two given nodes.
has_path_idx(a_idx, b_idx)	Checks if a path exists between the two given nodes.
iterate_parametrized_layers()	Parametrized layers of the circuit.
nodes_between(a, b)	Nodes on all the directed paths between the two given nodes.
print_contents()	Prints the contents of the quantum circuit.
serialize()	Serialize the quantum circuit graph based on the operations and observables in the circuit graph and the index of the variables used by them.
update_node(old, new)	Replaces the given circuit graph node with a new one.
wire_indices(wire)	Operator indices on the given wire.

ancestors(ops, sort=False)

Ancestors of a given set of operators.

Parameters:

• ops (Iterable[Operator]) – set of operators in the circuit

• sort=False (bool) – if True, sort the operators according

• index (to the topological order determined by the queue)

Returns:

ancestors of the given operators

Return type:

list[Operator]

ancestors_in_order(ops)

Operator ancestors in a topological order.

Currently the topological order is determined by the queue index.

Parameters:

ops (Iterable[Operator]) – set of operators in the circuit

Returns:

ancestors of the given operators, topologically ordered

Return type:

list[Operator]

ancestors_of_indexes(indexes, sort=False)

Ancestors of a given set of operators.

Parameters:

• indexes (Sequence[int]) – the index into the queue for the operator

• sort=False (bool) – if True, sort the operators according

• index (to the topological order determined by the queue)

Returns:

ancestors of the given operators

Return type:

list[Operator]

descendants(ops, sort=False)

Descendants of a given set of operators.

Parameters:

• ops (Iterable[Operator]) – set of operators in the circuit

• sort=False (bool) – if True, sort the operators according

• index (to the topological order determined by the queue)

Returns:

descendants of the given operators

Return type:

list[Operator]

descendants_in_order(ops)

Operator descendants in a topological order.

Currently the topological order is determined by the queue index.

Parameters:

ops (Iterable[Operator]) – set of operators in the circuit

Returns:

descendants of the given operators, topologically ordered

Return type:

list[Operator]

descendants_of_indexes(indexes, sort=False)

Descendants of a given set of operators.

Parameters:

• indexes (Sequence[int]) – the index into the queue for the operator

• sort=False (bool) – if True, sort the operators according

• index (to the topological order determined by the queue)

Returns:

descendants of the given operators

Return type:

list[Operator]

get_depth()

Depth of the quantum circuit (longest path in the DAG).

has_path(a, b)

Checks if a path exists between the two given nodes.

Parameters:

• a (Operator) – initial node

• b (Operator) – final node

Returns:

returns True if a path exists

Return type:

bool

has_path_idx(a_idx, b_idx)

Checks if a path exists between the two given nodes.

Parameters:

• a_idx (int) – initial node index

• b_idx (int) – final node index

Returns:

returns True if a path exists

Return type:

bool

iterate_parametrized_layers()

Parametrized layers of the circuit.

Returns:

layers with extra metadata

Return type:

Iterable[LayerData]

nodes_between(a, b)

Nodes on all the directed paths between the two given nodes.

Returns the set of all nodes s that fulfill \(a \le s \le b\). There is a directed path from a via s to b iff the set is nonempty. The endpoints belong to the path.

Parameters:

• a (Operator) – initial node

• b (Operator) – final node

Returns:

nodes on all the directed paths between a and b

Return type:

list[Operator]

print_contents()

Prints the contents of the quantum circuit.

serialize()

Serialize the quantum circuit graph based on the operations and observables in the circuit graph and the index of the variables used by them.

The string that is produced can be later hashed to assign a unique value to the circuit graph.

Returns:

serialized quantum circuit graph

Return type:

string

update_node(old, new)

Replaces the given circuit graph node with a new one.

Parameters:

• old (Operator) – node to replace

• new (Operator) – replacement

Raises:

ValueError – if the new Operator does not act on the same wires as the old one

wire_indices(wire)

Operator indices on the given wire.

Parameters:

wire (int) – wire to examine

Returns:

indices of operators on the wire, in temporal order

Return type:

list[int]

code/api/pennylane.CircuitGraph

 

Download Python script

 

Download Notebook

 

View on GitHub