# qiskit.providers.aer.library.SaveExpectationValue¶

class SaveExpectationValue(operator, label='expectation_value', unnormalized=False, pershot=False, conditional=False)[source]

Save expectation value of an operator.

Instruction to save the expectation value of a Hermitian operator.

The expectation value of a Hermitian operator $$H$$ for a simulator in quantum state :mathrhois given by $$\langle H\rangle = \mbox{Tr}[H.\rho]$$.

Paramètres
• operator (Pauli or SparsePauliOp or Operator) – a Hermitian operator.

• label (str) – the key for retrieving saved data from results.

• unnormalized (bool) – If True return save the unnormalized accumulated or conditional accumulated expectation value over all shot [Default: False].

• pershot (bool) – if True save a list of expectation values for each shot of the simulation rather than the average over all shots [Default: False].

• conditional (bool) – if True save the average or pershot data conditional on the current classical register values [Default: False].

Lève

ExtensionError – if the input operator is invalid or not Hermitian.

Note

This instruction can be directly appended to a circuit using the save_expectation_value() circuit method.

__init__(operator, label='expectation_value', unnormalized=False, pershot=False, conditional=False)[source]

Instruction to save the expectation value of a Hermitian operator.

The expectation value of a Hermitian operator $$H$$ for a simulator in quantum state :mathrhois given by $$\langle H\rangle = \mbox{Tr}[H.\rho]$$.

Paramètres
• operator (Pauli or SparsePauliOp or Operator) – a Hermitian operator.

• label (str) – the key for retrieving saved data from results.

• unnormalized (bool) – If True return save the unnormalized accumulated or conditional accumulated expectation value over all shot [Default: False].

• pershot (bool) – if True save a list of expectation values for each shot of the simulation rather than the average over all shots [Default: False].

• conditional (bool) – if True save the average or pershot data conditional on the current classical register values [Default: False].

Lève

ExtensionError – if the input operator is invalid or not Hermitian.

Note

This instruction can be directly appended to a circuit using the save_expectation_value() circuit method.

Methods

 __init__(operator[, label, unnormalized, …]) Instruction to save the expectation value of a Hermitian operator. add_decomposition(decomposition) Add a decomposition of the instruction to the SessionEquivalenceLibrary. Return the QasmQobjInstruction for the intructions. broadcast_arguments(qargs, cargs) Validation of the arguments. c_if(classical, val) Add classical condition on register or cbit classical and value val. copy([name]) Copy of the instruction. Special case. Return True .IFF. DEPRECATED: use instruction.reverse_ops(). Return a default OpenQASM string for the instruction. Creates an instruction with gate repeated n amount of times. For a composite instruction, reverse the order of sub-instructions. soft_compare(other) Soft comparison between gates. validate_parameter(parameter) Instruction parameters has no validation or normalization.

Attributes

 decompositions Get the decompositions of the instruction from the SessionEquivalenceLibrary. definition Return definition in terms of other basic gates. duration Get the duration. label Return instruction label params return instruction params. unit Get the time unit of duration.
add_decomposition(decomposition)

Add a decomposition of the instruction to the SessionEquivalenceLibrary.

assemble()

Return the QasmQobjInstruction for the intructions.

broadcast_arguments(qargs, cargs)

Validation of the arguments.

Paramètres
• qargs (List) – List of quantum bit arguments.

• cargs (List) – List of classical bit arguments.

Yields

Tuple(List, List) – A tuple with single arguments.

Lève

CircuitError – If the input is not valid. For example, the number of arguments does not match the gate expectation.

c_if(classical, val)

Add classical condition on register or cbit classical and value val.

copy(name=None)

Copy of the instruction.

Paramètres

name (str) – name to be given to the copied circuit, if None then the name stays the same.

Renvoie

a copy of the current instruction, with the name

updated if it was provided

Type renvoyé

qiskit.circuit.Instruction

property decompositions

Get the decompositions of the instruction from the SessionEquivalenceLibrary.

property definition

Return definition in terms of other basic gates.

property duration

Get the duration.

inverse()

Special case. Return self.

is_parameterized()

Return True .IFF. instruction is parameterized else False

property label

Return instruction label

Type renvoyé

str

mirror()

DEPRECATED: use instruction.reverse_ops().

Renvoie

a new instruction with sub-instructions

reversed.

Type renvoyé

qiskit.circuit.Instruction

property params

return instruction params.

qasm()

Return a default OpenQASM string for the instruction.

Derived instructions may override this to print in a different format (e.g. measure q[0] -> c[0];).

repeat(n)

Creates an instruction with gate repeated n amount of times.

Paramètres

n (int) – Number of times to repeat the instruction

Renvoie

Containing the definition.

Type renvoyé

qiskit.circuit.Instruction

Lève

CircuitError – If n < 1.

reverse_ops()

For a composite instruction, reverse the order of sub-instructions.

This is done by recursively reversing all sub-instructions. It does not invert any gate.

Renvoie

a new instruction with

sub-instructions reversed.

Type renvoyé

qiskit.circuit.Instruction

soft_compare(other)

Soft comparison between gates. Their names, number of qubits, and classical bit numbers must match. The number of parameters must match. Each parameter is compared. If one is a ParameterExpression then it is not taken into account.

Paramètres

other (instruction) – other instruction.

Renvoie

are self and other equal up to parameter expressions.

Type renvoyé

bool

property unit

Get the time unit of duration.

validate_parameter(parameter)

Instruction parameters has no validation or normalization.