qp.SumOfSlatersPrep¶
- class SumOfSlatersPrep(coefficients, wires, indices, enumeration_wires=(), identification_wires=(), qrom_work_wires=(), mcx_cache_wires=())[source]¶
Bases:
Operator2Prepare an arbitrary quantum state with the sum-of-Slaters technique.
This operation prepares an arbitrary state
\[|\psi\rangle = \sum_{\ell \in L } c_\ell |\ell\rangle,\]where \(L\) denotes the set of
indicesand \(c_\ell\) is thecoefficientcorresponding to the index \(\ell\in L\). The state \(|\ell\rangle\) is a computational basis state, interpreted via the binary representation of \(\ell\).This state preparation technique was introduced in Sec. III A of Fomichev et al., PRX Quantum 5, 040339 and is tailored to sparse states.
See also
select_sos_rows()andcompute_sos_encoding()for the required classical coprocessing, as well asPartialUnaryStatePreparationfor another sparse state preparation technique.- Parameters:
coefficients (np.ndarray) – Coefficients of the sparse state to prepare. The ordering should match that in
indices.wires (WiresLike) – Wires on which to prepare the state.
enumeration_wires (WiresLike) – Work wires used for the enumeration register. For \(d\) entries in the state, \(\lceil \log_2 (d)\rceil\) qubits are required.
identification_wires (WiresLike) – Work wires used for the identification register. The required number of qubits depends on the particular
indicesof the sparse state, but it is at most \(2d-1\) for \(d\) entries in the state.qrom_work_wires (WiresLike) – Work wires used by the
QROMsubroutine. For \(d\) entries in the state, \(\lceil \log_2 (d)\rceil - 1\) qubits are required.mcx_cache_wires (WiresLike) – Work wires used for caching AND values when uncomputing the enumeration register with multicontrolled bit flips. The required number of qubits depends on the particular
indicesof the sparse state, but it is at most \(2d-2\) for \(d\) entries in the state.indices (tuple[int] | None) – Indices of the sparse state to prepare. The ordering should match that in
coefficients.Noneindicates unknown indices for the purposes of resource estimation.
The sizes for the numerous optional work wire registers can be computed with
SumOfSlatersPrep.required_register_sizes(indices, wires).Warning
Note that we require
coefficientsto be treated as numerical data in the form of an array, whereas theindicesneed to be hashable, and thus will be treated as static information. This is becauseindicessignificantly impacts the structure and size of the circuit that realizes the state preparation.Example
Consider a sparse state specified by normalized coefficients and statevector indices pointing to the populated computational basis states:
import pennylane as qp import numpy as np coefficients = np.array([1, -1j, 1j, 1, 1, -1j, 1, 1j]) / np.sqrt(8) indices = (0, 1, 2, 4, 8, 16, 32, 64) wires = list(range(7))
This is all the information we require to create the state preparation:
coefficients,indices, andwires. Theindicescorrespond to the computational basis states interpreted via their binary representation (e.g., \(|3\rangle = |11\rangle\) for two qubits or \(|3\rangle = |011\rangle\) for three qubits).qp.decomposition.enable_graph() gate_set = {"QROM", "TemporaryAND", "Adjoint(TemporaryAND)", "MultiplexerStatePreparation", "CNOT", "X"} first_free_wire = max(wires)+1 @qp.transforms.resolve_dynamic_wires(min_int=first_free_wire) @qp.transforms._tape_decompose(gate_set=gate_set, num_work_wires=14) @qp.qnode(qp.device("lightning.qubit", wires=21)) def circuit(): qp.SumOfSlatersPrep(coefficients, wires, indices) return qp.state()
We can check that we prepared the right state:
>>> prepared_state = circuit()[::2**14] # Slice the state, as there are 14 work wires >>> where = np.where(prepared_state) >>> print(where) (array([ 0, 1, 2, 4, 8, 16, 32, 64]),) >>> # Adding 0.0 to the rounded result will prevent stochastic signed zeros >>> print(np.round(prepared_state[where], 4) + 0.0) [0.3536+0.j 0. -0.3536j 0. +0.3536j 0.3536+0.j 0.3536+0.j 0. -0.3536j 0.3536+0.j 0. +0.3536j]
That looks exactly right! Internally, the state preparation looks like this:
>>> print(qp.draw(circuit, show_matrices=False, max_length=195)()) 0: ──────────────────────────────────╭QROM(M1)─╭●───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 1: ──────────────────────────────────├QROM(M1)─│────────╭●──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 2: ──────────────────────────────────├QROM(M1)─│──╭●────│──╭●───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 3: ──────────────────────────────────├QROM(M1)─│──│─────│──│─────╭●─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 4: ──────────────────────────────────├QROM(M1)─│──│──╭●─│──│──╭●─│──╭●──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 5: ──────────────────────────────────├QROM(M1)─│──│──│──│──│──│──│──│──╭●───────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 6: ──────────────────────────────────├QROM(M1)─│──│──│──│──│──│──│──│──│──╭●────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 7: ─╭MultiplexerStatePreparation(M0)─├◑────────│──│──│──│──│──│──│──│──│──│─────────────────────────────────────────────────────────────────────────────────────────────────────────────────── ··· 8: ─├MultiplexerStatePreparation(M0)─├◑────────│──│──│──│──│──│──│──│──│──│───────────────────────────────────────────────────╭X────────────────────────────────╭X──────────────────────────── ··· 9: ─╰MultiplexerStatePreparation(M0)─├◑────────│──│──│──│──│──│──│──│──│──│─────────────────╭X────────────────────────────────│─────────────────────────────────│──╭X───────────────────────── ··· 10: ──────────────────────────────────│─────────╰X─╰X─╰X─│──│──│──│──│──│──│───X─╭●──────────│───────────────●╮────╭●──────────│───────────────●╮──X─╭●──────────│──│───────────────●╮──X─╭●─── ··· 11: ──────────────────────────────────│──────────────────╰X─╰X─╰X─│──│──│──│───X─├●──────────│───────────────●┤────├●──────────│───────────────●┤──X─├●──────────│──│───────────────●┤──X─├●─── ··· 12: ──────────────────────────────────│───────────────────────────╰X─╰X─│──│───X─│──╭●───────│───────────●╮───│────│──╭●───────│───────────●╮───│──X─│──╭●───────│──│───────────●╮───│────│──╭● ··· 13: ──────────────────────────────────│─────────────────────────────────╰X─│───X─│──│──╭●────│───────●╮───│───│──X─│──│──╭●────│───────●╮───│───│──X─│──│──╭●────│──│───────●╮───│───│────│──│─ ··· 14: ──────────────────────────────────│────────────────────────────────────╰X────│──│──│──╭●─│───●╮───│───│───│──X─│──│──│──╭●─│───●╮───│───│───│────│──│──│──╭●─│──│───●╮───│───│───│────│──│─ ··· 15: ──────────────────────────────────├work──────────────────────────────────────│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│─ ··· 16: ──────────────────────────────────╰work──────────────────────────────────────│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│─ ··· 17: ─────────────────────────────────────────────────────────────────────────────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────╰⊕─├● ··· 18: ────────────────────────────────────────────────────────────────────────────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯───────────╰⊕ ··· 19: ───────────────────────────────────────────────────────────────────────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯───────────────── ··· 20: ──────────────────────────────────────────────────────────────────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯───────────────────── ··· 0: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╭●────────────────────────────┤ ╭State 1: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│────────╭●───────────────────┤ ├State 2: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│──╭●────│──╭●────────────────┤ ├State 3: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│──│─────│──│─────╭●──────────┤ ├State 4: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│──│──╭●─│──│──╭●─│──╭●───────┤ ├State 5: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│──│──│──│──│──│──│──│──╭●────┤ ├State 6: ··· ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────│──│──│──│──│──│──│──│──│──╭●─┤ ├State 7: ··· ───────╭X────────────────────────────────╭X───────────────────────────────────╭X───────────────────────────────────╭X───────────────────────│──│──│──│──│──│──│──│──│──│──┤ ├State 8: ··· ───────│─────────────────────────────────│────────────────────────────────────│──╭X────────────────────────────────│──╭X────────────────────│──│──│──│──│──│──│──│──│──│──┤ ├State 9: ··· ───────│─────────────────────────────────│──╭X────────────────────────────────│──│─────────────────────────────────│──│──╭X─────────────────│──│──│──│──│──│──│──│──│──│──┤ ├State 10: ··· ───────│───────────────●╮──X─╭●──────────│──│───────────────●╮──X─╭●──────────│──│───────────────●╮──X─╭●──────────│──│──│───────────────●╮─╰X─╰X─╰X─│──│──│──│──│──│──│──┤ ├State 11: ··· ───────│───────────────●┤──X─├●──────────│──│───────────────●┤────├●──────────│──│───────────────●┤──X─├●──────────│──│──│───────────────●┤──X───────╰X─╰X─╰X─│──│──│──│──┤ ├State 12: ··· ───────│───────────●╮───│──X─│──╭●───────│──│───────────●╮───│────│──╭●───────│──│───────────●╮───│────│──╭●───────│──│──│───────────●╮───│──X────────────────╰X─╰X─│──│──┤ ├State 13: ··· ─╭●────│───────●╮───│───│────│──│──╭●────│──│───────●╮───│───│────│──│──╭●────│──│───────●╮───│───│────│──│──╭●────│──│──│───────●╮───│───│──X──────────────────────╰X─│──┤ ├State 14: ··· ─│──╭●─│───●╮───│───│───│────│──│──│──╭●─│──│───●╮───│───│───│────│──│──│──╭●─│──│───●╮───│───│───│────│──│──│──╭●─│──│──│───●╮───│───│───│──X─────────────────────────╰X─┤ ├State 15: ··· ─│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──│────│───│───│───│───────────────────────────────┤ ├State 16: ··· ─│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──│────│───│───│───│───────────────────────────────┤ ├State 17: ··· ─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│──│────│───│──●┤──⊕╯───────────────────────────────┤ ├State 18: ··· ─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│──│────│──●┤──⊕╯───────────────────────────────────┤ ├State 19: ··· ─╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯──────────────────╰⊕─├●─│──│──│───●┤──⊕╯───────────────────────────────────────┤ ├State 20: ··· ────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●─╰●──⊕╯───────────────────────────────────────────┤ ╰State
Here, the first seven wires (
0to6) are the target wires of the state preparation, wires7, 8, 9form the enumeration register, the next five wires (10to14) are the encoding register, and the pair of wires15, 16as well as the wires17to20are work wires for theQROMand the enumeration uncomputation, respectively.Available decompositions
Resources are general, for \(D \geq 2\) entries in
indices, \(d = \lceil \log_2 D \rceil\) enumeration wires and \(m\) encoding qubits (the number of rows of the encoding computed bycompute_sos_encoding()). Each circuit shows one example of the rule. Use a rule when decomposing withqp.decompose:qp.decompose(gate_set=..., fixed_decomps={qp.SumOfSlatersPrep: "<name>"})
_sos_state_prepExample:
indices=(0, 1, 2, 4)on three wires, \(D=4\), \(m=3\), \(u=0\)Gate
Count
Example
MultiplexerStatePreparationon \(d\) wires\(1\)
1
\(1\)
1
\(2u + \sum_{k=1}^{D-1} \mathrm{wt}(k)\)
4
\((D-1)(m-1)\)
6
\((D-1)(m-1)\)
6
\(N_X \leq D m\)
8
where \(u\) is the number of ones in the encoding matrix (\(u=0\) for an identity encoding), \(\mathrm{wt}(k)\) the number of ones in the binary representation of \(k\), and \(N_X\) the number of bit flips between consecutive identification bit strings (data dependent)
Usage details
Reduced circuit complexity for identity encodings
Depending on the
indicespassed to the state preparation, they may or may not be reducible to short enough sub-bitstrings such that no further encoding is required. In this case, the blocks ofCNOTgates after theQROMand at the end, as seen in the example above, are not needed. For example, consider the following modification of the example:coefficients = np.array([0.25, 0.25j, -0.25, 0.5, 0.5, 0.25, -0.25j, 0.25, -0.25, 0.25]) indices = (0, 1, 4, 13, 14, 17, 19, 22, 23, 25) wires = list(range(5)) first_free_wire = max(wires)+1 @qp.transforms.resolve_dynamic_wires(min_int=first_free_wire) @qp.transforms._tape_decompose(gate_set=gate_set, num_work_wires=11) @qp.qnode(qp.device("lightning.qubit", wires=16)) def circuit(): qp.SumOfSlatersPrep(coefficients, wires, indices) return qp.state()
In this case, we only require eight work wires, because the encoding blocks can be skipped.
>>> prepared_state = circuit()[::2**11] # Slice the state, as there are eleven work wires >>> where = np.where(np.abs(prepared_state)>1e-12) >>> print(where) (array([ 0, 1, 4, 13, 14, 17, 19, 22, 23, 25]),) >>> # Adding 0.0 to the rounded result will prevent stochastic signed zeros >>> print(np.round(prepared_state[where], 4) + 0.0) [ 0.25+0.j 0. +0.25j -0.25+0.j 0.5 +0.j 0.5 +0.j 0.25+0.j 0. -0.25j 0.25+0.j -0.25+0.j 0.25+0.j ]
The reduced circuit looks like this:
>>> print(qp.draw(circuit, show_matrices=False, max_length=205)()) 0: ──────────────────────────────────╭QROM(M1)──X─╭●──────────────────────────●╮────╭●──────────────────────────●╮────╭●─────────────────────────────●╮────╭●──────────────────────────●╮──X─╭●───────── ··· 1: ──────────────────────────────────├QROM(M1)──X─├●──────────────────────────●┤────├●──────────────────────────●┤──X─├●─────────────────────────────●┤────├●──────────────────────────●┤──X─├●───────── ··· 2: ──────────────────────────────────├QROM(M1)──X─│──╭●───────────────────●╮───│──X─│──╭●───────────────────●╮───│────│──╭●──────────────────────●╮───│────│──╭●───────────────────●╮───│──X─│──╭●────── ··· 3: ──────────────────────────────────├QROM(M1)──X─│──│──╭●────────────●╮───│───│────│──│──╭●────────────●╮───│───│────│──│──╭●───────────────●╮───│───│──X─│──│──╭●────────────●╮───│───│──X─│──│──╭●─── ··· 4: ──────────────────────────────────├QROM(M1)────│──│──│──╭●─────●╮───│───│───│──X─│──│──│──╭●─────●╮───│───│───│──X─│──│──│──╭●────────●╮───│───│───│──X─│──│──│──╭●─────●╮───│───│───│──X─│──│──│──╭● ··· 5: ─╭MultiplexerStatePreparation(M0)─├◑───────────│──│──│──│───────│───│───│───│────│──│──│──│───────│───│───│───│────│──│──│──│──────────│───│───│───│────│──│──│──│───────│───│───│───│────│──│──│──│─ ··· 6: ─├MultiplexerStatePreparation(M0)─├◑───────────│──│──│──│───────│───│───│───│────│──│──│──│───────│───│───│───│────│──│──│──│──────────│───│───│───│────│──│──│──│──╭X───│───│───│───│────│──│──│──│─ ··· 7: ─├MultiplexerStatePreparation(M0)─├◑───────────│──│──│──│───────│───│───│───│────│──│──│──│──╭X───│───│───│───│────│──│──│──│──╭X──────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│─ ··· 8: ─╰MultiplexerStatePreparation(M0)─├◑───────────│──│──│──│──╭X───│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──╭X───│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│─ ··· 9: ──────────────────────────────────├work────────│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│─ ··· 10: ──────────────────────────────────├work────────│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│─ ··· 11: ──────────────────────────────────╰work────────│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│─ ··· 12: ───────────────────────────────────────────────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│─ ··· 13: ──────────────────────────────────────────────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│─ ··· 14: ─────────────────────────────────────────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯──────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├● ··· 15: ────────────────────────────────────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕ ··· 0: ··· ────────────────────●╮────╭●─────────────────────────────●╮────╭●────────────────────────────────●╮────╭●──────────────────────────●╮────╭●─────────────────────────────●╮────┤ ╭State 1: ··· ────────────────────●┤────├●─────────────────────────────●┤────├●────────────────────────────────●┤────├●──────────────────────────●┤──X─├●─────────────────────────────●┤────┤ ├State 2: ··· ────────────────●╮───│────│──╭●──────────────────────●╮───│──X─│──╭●─────────────────────────●╮───│────│──╭●───────────────────●╮───│──X─│──╭●──────────────────────●╮───│──X─┤ ├State 3: ··· ────────────●╮───│───│──X─│──│──╭●───────────────●╮───│───│────│──│──╭●──────────────────●╮───│───│────│──│──╭●────────────●╮───│───│──X─│──│──╭●───────────────●╮───│───│──X─┤ ├State 4: ··· ────────●╮───│───│───│────│──│──│──╭●────────●╮───│───│───│──X─│──│──│──╭●───────────●╮───│───│───│──X─│──│──│──╭●─────●╮───│───│───│────│──│──│──╭●────────●╮───│───│───│────┤ ├State 5: ··· ─────────│───│───│───│────│──│──│──│──────────│───│───│───│────│──│──│──│─────────────│───│───│───│────│──│──│──│──╭X───│───│───│───│────│──│──│──│──╭X──────│───│───│───│────┤ ├State 6: ··· ─╭X──────│───│───│───│────│──│──│──│──╭X──────│───│───│───│────│──│──│──│──╭X─────────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│───────│───│───│───│────┤ ├State 7: ··· ─│───────│───│───│───│────│──│──│──│──│──╭X───│───│───│───│────│──│──│──│──│──╭X──────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│───────│───│───│───│────┤ ├State 8: ··· ─│──╭X───│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──╭X───│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──╭X───│───│───│───│────┤ ├State 9: ··· ─│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────┤ ├State 10: ··· ─│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────┤ ├State 11: ··· ─│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────│──│──│──│──│──│──│────│───│───│───│────│──│──│──│──│────│───│───│───│────│──│──│──│──│──│────│───│───│───│────┤ ├State 12: ··· ─│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│────│───│──●┤──⊕╯────╰⊕─├●─│──│──│──│────│───│──●┤──⊕╯────┤ ├State 13: ··· ─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│────│──●┤──⊕╯───────────╰⊕─├●─│──│──│────│──●┤──⊕╯────────┤ ├State 14: ··· ─│──│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯──────────────────╰⊕─├●─│──│──│───●┤──⊕╯──────────────────╰⊕─├●─│───●┤──⊕╯──────────────────╰⊕─├●─│──│───●┤──⊕╯────────────┤ ├State 15: ··· ─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●─╰●──⊕╯─────────────────────────╰⊕─╰●──⊕╯─────────────────────────╰⊕─╰●─╰●──⊕╯────────────────┤ ╰State
As we can see, the ladders of elbow, or
TemporaryANDgates are now controlled on the target register directly, rather than the encoding register, which we thus can skip for the identity encoding.Dynamic work wires
Note that in the example above, wires with labels
5to15were dynamically allocated. These registers can optionally be provided explicitly. If they are not, they will be dynamically allocated. We can see an initial dense state preparation viaStatePrepon fewer qubits (depicted as|Ψ⟩on the first four dynamic wires in the above diagram), aQROMand a sequence of elbow ladders that set a caching qubit (qubit index15), which then controlsCNOTgates that perform the actual uncomputation.Note that we guessed the required number of work wires (
num_work_wires) indecompose()and employedresolve_dynamic_wires()to assign integer wire labels to those dynamically allocated wires. If we want to know the required wire register sizes ahead of time, they can be computed withSumOfSlatersPrep.required_register_sizes:>>> sizes = qp.SumOfSlatersPrep.required_register_sizes(indices, len(wires)) >>> print(sizes) {'wires': 5, 'enumeration_wires': 4, 'identification_wires': 0, 'qrom_work_wires': 3, 'mcx_cache_wires': 4}
Note that work wires of subroutines like
QROMor the elbow ladders that realize aMultiControlledXgate are accounted for explicitly. Even thought the decomposition ofSumOfSlatersPrepallocates the needed work wires automatically, you may want to allocate them explicitly. Generically, this can be done withregisters():>>> print(qp.registers(sizes)) {'wires': Wires([0, 1, 2, 3, 4]), 'enumeration_wires': Wires([5, 6, 7, 8]), 'identification_wires': Wires([]), 'qrom_work_wires': Wires([9, 10, 11]), 'mcx_cache_wires': Wires([12, 13, 14, 15])}
Attributes
The expected types for the arguments of an operator.
Dictionary mapping argument names to their values.
Arithmetic depth of the operator.
Batch size of the operator if it is used with broadcasted parameters.
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
The names of arguments that are treated as compilable static arguments.
Dictionary mapping compilable argument names to their values.
Legacy Operator compatibility view of dynamic numerical arguments.
The names of arguments that are treated as dynamic.
Dictionary mapping dynamic argument names to their values.
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
Gradient computation method.
Compute 'grad_recipe' lazily.
Whether the expected signature of an operator is fixed.
The names of arguments that represent dynamic data wrapped in static structures (known as Pytrees).
Dictionary mapping hybrid argument names to their values.
Legacy Operator compatibility view of non-dynamic, non-wire constructor arguments.
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
Whether this operator contains only abstract data.
This property determines if an operator is verified to be Hermitian.
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
Operator name.
Number of dimensions per trainable parameter of the operator.
Number of trainable parameters.
Legacy Operator compatibility view of dynamic numerical arguments as a list.
A
PauliSentencerepresentation of the Operator, orNoneif it doesn't have one.partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
The names of arguments that are treated as static.
Dictionary mapping static argument names to their values.
The names of arguments corresponding to wires.
Dictionary mapping wire argument names to their values.
The expected number of wire labels for each wire argument.
Wires that the operator acts on.
- arg_specs = {'wires': AbstractWires(-1)}¶
The expected types for the arguments of an operator. This attribute is optional—not setting it has no loss of functionality. If set, it can be used to perform automatic validation of an operators inputs during construction. Additionally, when defining decomposition rules for an operator, operator types with
arg_specsthat spans all the arguments with static types can be placed in the rules’ resources without needing to fully construct abstract operators.Note
A type that is listed in ‘arg_specs’ says what an argument is allowed to be, not what it actually is. For example, if arg_specs contains Complex[-1, -1], the Operator can still be instantiated with a real float64 array, which will then be reported as complex even though it holds real data.
The decomposition graph goes by the reported type, so real and complex inputs will look like the same operator and share one rule. To let them decompose differently, leave the argument out of
arg_specsand give each rule aregister_conditionthat checks the type. For a concrete example seeBasisRotation.
- arguments¶
Dictionary mapping argument names to their values.
- arithmetic_depth¶
Arithmetic depth of the operator.
- batch_size¶
Batch size of the operator if it is used with broadcasted parameters.
The
batch_sizeis determined based onndim_paramsand 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
- coefficients¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- compilable_argnames = ('indices',)¶
The names of arguments that are treated as compilable static arguments. Compilable static arguments are a subset of static arguments—arguments whose concrete values are known when compiling the program. But, unlike
static_argnames, they are lowered to the compiler intermediate representation, making their values visible to the compiler, which may be useful if the compiler needs to interact with such values. Such values may include numbers, strings, and lists, tuples, or dictionaries thereof. This feature is opt-in; if any static arguments are not guaranteed to be compilable, it is safer to place them instatic_argnames.Note
An operator can only specify
static_argnamesandhybrid_argnames, orcompilable_argnames, but not both; if any static or hybrid arguments are not or cannot be lowered to the IR, then all static and hybrid arguments are assumed to not be lowerable.
- compilable_args¶
Dictionary mapping compilable argument names to their values.
- data¶
Legacy Operator compatibility view of dynamic numerical arguments.
- dynamic_argnames = ('coefficients',)¶
The names of arguments that are treated as dynamic. Dynamic arguments are numerical arguments whose concrete values may not be known at compile-time, such as rotation angles, matrices, etc., and may include trainable data. Additionally, dynamic arguments are the only ones used to infer
batch_size, necessary for broadcasting support.Inputs for these arguments must be scalars, arrays, or castable to arrays (e.g. multi- dimensional lists with homogenous shapes), and operator constructors will be responsible for “canonicalizing” them (such as casting a list input to an array). At compile-time, the concrete values of these arguments are assumed to be unknown—only their shapes and data types are known.
- dynamic_args¶
Dictionary mapping dynamic argument names to their values.
- enumeration_wires¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- grad_method¶
Gradient computation method.
'A': analytic differentiation using the parameter-shift method.'F': finite difference numerical differentiation.None: the operation may not be differentiated.
Default is
'F', orNoneif the Operation has zero parameters.
- grad_recipe¶
Compute ‘grad_recipe’ lazily.
- has_adjoint = False¶
- has_decomposition = True¶
- has_diagonalizing_gates = False¶
- has_fixed_sig = False¶
Whether the expected signature of an operator is fixed. If
True, then the operator’s signature will always be fully known. When defining decomposition rules for an operator, operator types with fixed signatures can be placed in the rules’ resources without needing to fully construct abstract operators. This is set automatically whenarg_specscovers every dynamic and wire argument, there are no hybrid, static, or compilable arguments, and every declared type is fully fixed (no unknown array shapes or wire counts).
- has_generator = False¶
- has_matrix = False¶
- has_sparse_matrix = False¶
- hybrid_argnames = ()¶
The names of arguments that represent dynamic data wrapped in static structures (known as Pytrees). Names in this category must be disjoint from
dynamic_argnames,static_argnames, andcompilable_argnames, but may overlap withwire_argnameswhen those arguments contain nested structures of wires. Examples of hybrid arguments include collections of wires or dynamic arrays, operators, etc.Note
An operator can only specify
static_argnamesandhybrid_argnames, orcompilable_argnames, but not both; if any static or hybrid arguments are not or cannot be lowered to the IR, then all static and hybrid arguments are assumed to not be lowerable.
- hybrid_args¶
Dictionary mapping hybrid argument names to their values.
- hyperparameters¶
Legacy Operator compatibility view of non-dynamic, non-wire constructor arguments.
- identification_wires¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- indices¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- is_fully_abstract¶
Whether this operator contains only abstract data.
- is_verified_hermitian¶
This property determines if an operator is verified to be Hermitian.
Note
This property provides a fast, non-exhaustive check used for internal optimizations. It relies on quick, provable shortcuts (e.g., operator properties) rather than a full, computationally expensive check.
For a definitive check, use the
pennylane.is_hermitian()function. Please note that this comes with increased computational cost.- Returns:
The property will return
Trueif the operator is guaranteed to be Hermitian andFalseif the check is inconclusive and the operator may or may not be Hermitian.- Return type:
bool
Consider this operator,
>>> op = (qp.X(0) @ qp.Y(0) - qp.X(0) @ qp.Z(0)) * 1j
In this case, Hermicity cannot be verified and leads to an inconclusive result:
>>> op.is_verified_hermitian # inconclusive False
However, using
pennylane.is_hermitian()will give the correct answer:>>> qp.is_hermitian(op) # definitive True
- mcx_cache_wires¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- name¶
Operator name.
- 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.
- parameters¶
Legacy Operator compatibility view of dynamic numerical arguments as a list.
- pauli_rep¶
A
PauliSentencerepresentation of the Operator, orNoneif it doesn’t have one.
- qrom_work_wires¶
partial(func, *args, **keywords) - new function with partial application of the given arguments and keywords.
- static_argnames = ()¶
The names of arguments that are treated as static. Static arguments are those whose concrete values are known when compiling the program. Arguments specified here are not lowered to a compiler intermediate representation (IR). Alternatively, if an argument is of a type that can be lowered, it can be moved to
compilable_argnames.Note
An operator can only specify
static_argnamesandhybrid_argnames, orcompilable_argnames, but not both; if any static or hybrid arguments are not or cannot be lowered to the IR, then all static and hybrid arguments are assumed to not be lowerable.
- static_args¶
Dictionary mapping static argument names to their values.
- wire_argnames = ('wires', 'enumeration_wires', 'identification_wires', 'qrom_work_wires', 'mcx_cache_wires')¶
The names of arguments corresponding to wires. Values for these arguments are automatically wrapped in
Wiresobjects by theOperator2constructor. If an argument is expected to be a structure that wraps wires (known as pytrees), such as a list of wire registers, make sure to include its name inhybrid_argnamesas well. Additionally,Operator2does not descend into the pytree structure of such arguments, so subclasses must ensure every wire leaf inside a hybrid wire argument is already aWiresobject before forwarding it tosuper().__init__.The order in which names appear in
wire_argnamesdetermines the order in which their wires appear inop.wires(seeOperator2.wires). For hybrid wire arguments, the containedWiresleaves are ordered by pytree traversal order. Wires contributed by arguments which are themselvesOperator2objects are appended after allwire_argnameswires. The special names"work_wires"and"work_wire"may be included inwire_argnamesbut their values are excluded fromop.wires.
- wire_args¶
Dictionary mapping wire argument names to their values.
- wire_sizes = (None, None, None, None, None)¶
The expected number of wire labels for each wire argument. If any wire arguments support an arbitrary size,
Nonemust be used. By default, all wire arguments are assumed to support arbitrary sizes. For hybrid wire arguments, the wire size must always beNone. Specifying wire sizes allows for additional automatic validation to be implemented, but, specifying it is optional if such validation is not needed.
- wires¶
Wires that the operator acts on.
By default, the returned
Wiresare collected from the operator’s arguments in the following order:For each name in
wire_argnames(in declaration order):If the name is not in
hybrid_argnames, the canonical value of that argument is added.If the name is in
hybrid_argnames, everyWiresleaf of the argument’s pytree is added in pytree traversal order.
After all
wire_argnameshave been processed, for each name inhybrid_argnamesthat is not inwire_argnames, the wires of anyOperator2leaves inside that argument are appended to the end in pytree traversal order.
Duplicate wires are removed while preserving the first occurrence, so the final result is the ordered union of all wires above.
Note
By default, work wires are not included in
op.wires. In particular, wire arguments namedwork_wiresorwork_wireare excluded.Note
This property may be overridden by developers if the default behaviour is not desired, for reasons such as including work wires or changing the order of the wires.
- Returns:
wires
- Return type:
Methods
adjoint()Create an operation that is the adjoint of this one.
compute_decomposition(*args, **kwargs)Representation of the operator as a product of other operators (static method).
compute_diagonalizing_gates(*args, **kwargs)Sequence of gates that diagonalize the operator in the computational basis (static method).
compute_eigvals(*args, **kwargs)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(*args[, format])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 new 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.
required_register_sizes(indices, num_wires)Compute the register sizes required for
SumOfSlatersPrep, for given computational basis states,indices, and number of target wires,num_wires.simplify()Reduce the depth of nested operators to the minimum.
sparse_matrix([wire_order, format])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. Used to simplify
Adjointoperators constructed byadjoint(). TODO: [sc-120844] Fix docstring after Adjoint is addedAdjointed operations are the conjugated and transposed version of the original operation. Adjointed ops are equivalent to the inverted operation for unitary gates.
Operator2.adjointcan be optionally defined by Operator developers, whileadjoint()is the entry point for constructing generic adjoint representations.- Returns:
The adjointed operation.
>>> from pennylane.core.operator import Operator2 >>> class MyClass(Operator2): ... dynamic_argnames = ("phi",) ... ... def __init__(self, phi, wires): ... super().__init__(phi, wires) ... ... def adjoint(self): ... return self ... >>> op = MyClass(0.5, wires=0).adjoint() >>> op MyClass(0.5, wires=[0])
- classmethod compute_decomposition(*args, **kwargs)¶
Representation of the operator as a product of other operators (static method).
\[O = O_1 O_2 \dots O_n.\]Note
Operations making up the decomposition should be queued within the
compute_decompositionmethod.See also
- Parameters:
*args (tuple) – Positional arguments of the operator
**kwargs (dict) – Keyword arguments of the operator
- Returns:
decomposition of the operator
- Return type:
list[Operator2]
- static compute_diagonalizing_gates(*args, **kwargs)¶
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:
*args (tuple) – Positional arguments of the operator
**kwargs (dict) – Keyword arguments of the operator
- Returns:
list of diagonalizing gates
- Return type:
list[.Operator2]
- static compute_eigvals(*args, **kwargs)¶
Eigenvalues of the operator in the computational basis (static method).
If
diagonalizing_gatesare 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:
*args (tuple) – Positional arguments of the operator
**kwargs (dict) – Keyword arguments of the operator
- Returns:
eigenvalues
- Return type:
tensor_like
- static compute_matrix(*args, **kwargs)¶
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.
See also
- Parameters:
*args (tuple) – Positional arguments of the operator
**kwargs (dict) – Keyword arguments of the operator
- Returns:
matrix representation
- Return type:
tensor_like
- static compute_sparse_matrix(*args, format='csr', **kwargs)¶
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:
*args (tuple) – Positional arguments of the operator
format (str) – format of the returned scipy sparse matrix, for example ‘csr’
**kwargs (dict) – Keyword arguments of the operator
- 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
DecompositionUndefinedErroris raised if no representation by decomposition is defined.See also
- Returns:
decomposition of the operator
- Return type:
list[Operator2]
- 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
DiagGatesUndefinedErroris raised if no representation by decomposition is defined.See also
- Returns:
a list of operators
- Return type:
list[.Operator2] or None
- eigvals()¶
Eigenvalues of the operator in the computational basis.
If
diagonalizing_gatesare 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
EigvalsUndefinedErroris 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
LinearCombinationandSparseHamiltonianrespectively).
- label(decimals=None, base_label=None, cache=None)¶
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 = qp.RX(1.23456, wires=0) >>> op.label() 'RX' >>> op.label(base_label="my_label") 'my_label' >>> op = qp.RX(1.23456, wires=0) >>> op.label() 'RX' >>> op.label(decimals=2) 'RX\n(1.23)' >>> op.label(base_label="my_label") 'my_label' >>> op.label(decimals=2, base_label="my_label") 'my_label\n(1.23)'
Note that by default, operator wires and static inputs are not included in the label. To override this behaviour,
Operator2subclasses must override thelabelmethod.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 = qp.QubitUnitary(np.eye(2), wires=0) >>> cache = {'matrices': []} >>> op2.label(cache=cache) 'U\n(M0)' >>> cache['matrices'] [array([[1., 0.], [0., 1.]])] >>> op3 = qp.QubitUnitary(np.eye(4), wires=(0,1)) >>> op3.label(cache=cache) 'U\n(M1)' >>> cache['matrices'] [array([[1., 0.], [0., 1.]]), array([[1., 0., 0., 0.], [0., 1., 0., 0.], [0., 0., 1., 0.], [0., 0., 0., 1.]])]
- map_wires(wire_map)¶
Returns a new 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:
.Operator2
- matrix(wire_order=None)¶
Representation of the operator as a matrix in the computational basis.
If
wire_orderis 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
MatrixUndefinedErroris 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. This method is used to simplify
Powinstances created bypow()orop ** power. TODO: [sc-120843] Fix docstring after Pow is addedOperator2.powcan be optionally defined by Operator developers, whilepow()orop ** powerare the entry point for constructing generic powers to exponents.- Parameters:
z (float) – exponent for the operator
- Returns:
list[
Operator2]
>>> from pennylane.core.operator import Operator2 >>> class MyClass(Operator2): ... dynamic_argnames = ("phi",) ... ... def __init__(self, phi, wires): ... super().__init__(phi, wires) ... ... def pow(self, z): ... return [MyClass(self.phi*z, self.wires)] ... >>> MyClass(0.5, 0).pow(2) [MyClass(1.0, wires=[0])]
- queue(context=<class 'pennylane.core.queuing.QueuingManager'>)¶
Append the operator to the Operator queue.
- static required_register_sizes(indices, num_wires)[source]¶
Compute the register sizes required for
SumOfSlatersPrep, for given computational basis states,indices, and number of target wires,num_wires.- Parameters:
indices (tuple[int] or AbstractArray) – Indices of computational basis states of the sparse state to be prepared with
SumOfSlatersPrep.num_wires (int) – Number of target wires on which
SumOfSlatersPrepwill prepare the state.
- Returns:
Required register size per register name. Includes the target wires of length
num_wireswith the label"wires", matching the call signature ofSumOfSlatersPrep.- Return type:
dict[str, int]
This function supports an abstract input for
indices, in which case an upper bound for the register sizes, across all possible sets of basis states of the given length, is returned:>>> indices = qp.typing.Int[45] >>> qp.SumOfSlatersPrep.required_register_sizes(indices, 18) {'wires': 18, 'enumeration_wires': 6, 'identification_wires': 11, 'qrom_work_wires': 5, 'mcx_cache_wires': 10}
- simplify()¶
Reduce the depth of nested operators to the minimum.
- Returns:
simplified operator
- Return type:
.Operator2
- sparse_matrix(wire_order=None, format='csr')¶
Representation of the operator as a sparse matrix in the computational basis.
If
wire_orderis 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
SparseMatrixUndefinedErroris 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
format (str) – format of the returned scipy sparse matrix, for example ‘csr’
- 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
TermsUndefinedErroris 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[.Operator2]]