Source code for qemcmc.circuits

import pennylane as qml
import numpy as np
from qemcmc.model.energy_model import EnergyModel
from typing import List


[docs] class CircuitMaker: """ Constructs and simulates quantum circuits used to generate QeMCMC proposals. This class builds the Hamiltonian corresponding to a given energy model and simulates its time evolution using PennyLane. Starting from a classical bitstring configuration, the circuit performs Trotterised quantum evolution and samples a new configuration from the resulting quantum state. The generated sample serves as the proposal state in the quantum-enhanced MCMC algorithm. Parameters ---------- model : EnergyModel Energy model defining the problem Hamiltonian. Notes ----- The total Hamiltonian simulated by the circuit is ``H = γ H_mixer + (1 - γ) α H_problem`` where ``H_problem`` encodes the classical energy model and ``H_mixer`` corresponds to a transverse-field term. The evolution time ``t`` and number of Trotter steps ``r`` are passed per call, giving an effective Trotter step size of ``δt = t / r``. The evolution is approximated using Trotterisation via ``qml.ApproxTimeEvolution``. """ def __init__(self, model: EnergyModel):
[docs] self.model = model
[docs] self.n_qubits = model.n
[docs] self.dev = qml.device("lightning.qubit", wires=self.n_qubits)
# self.dev = qml.device("default.tensor", wires=self.n_qubits, method="mps", max_bond_dim=2, contract="auto-mps")
[docs] self.model_type = model.model_type
[docs] self.devices = {} # cache devices for dynamic subgroup sizes if needed
[docs] def _get_device(self, num_wires: int): """Get or create a PennyLane device for the given number of wires.""" if num_wires not in self.devices: self.devices[num_wires] = qml.device("lightning.qubit", wires=num_wires) # self.devices[num_wires] = qml.device("default.tensor", wires=num_wires, method="mps", max_bond_dim=2, contract="auto-mps") return self.devices[num_wires]
[docs] def get_problem_hamiltonian(self, couplings: List[np.ndarray], sign: int = 1) -> qml.Hamiltonian: """ Construct the problem Hamiltonian from symmetric coupling tensors. This method supports both 'ising' (-1/+1) and 'binary' (0/1) models. Parameters ---------- couplings : List[np.ndarray] A list of coupling tensors. sign : int, optional A sign to apply to the Hamiltonian. Default is ``1``. Returns ------- qml.Hamiltonian The problem Hamiltonian. """ total_hamiltonian = 0.0 * qml.Identity(0) for coupling_tensor in couplings: coupling_tensor = np.asarray(coupling_tensor) order = coupling_tensor.ndim if order == 0: continue spin_sign = (-1) ** order if self.model_type == "ising" else 1 non_zero_indices = np.transpose(np.nonzero(coupling_tensor)) for index_tuple in non_zero_indices: index_tuple = tuple(int(i) for i in index_tuple) if len(set(index_tuple)) != len(index_tuple): # skip repeated indices continue if index_tuple != tuple(sorted(index_tuple)): # keep only strictly increasing i1 < i2 < ... < ik continue coeff = float(coupling_tensor[index_tuple]) if coeff == 0.0: continue if self.model_type == "ising": term = qml.PauliZ(index_tuple[0]) for q in index_tuple[1:]: term = term @ qml.PauliZ(q) total_hamiltonian += (sign * spin_sign * coeff) * term elif self.model_type == "binary": # 0.5 * (I - Z) for first variable term = 0.5 * (qml.Identity(index_tuple[0]) - qml.PauliZ(index_tuple[0])) # multiply by 0.5 * (I - Z) for rest for q in index_tuple[1:]: next_var = 0.5 * (qml.Identity(q) - qml.PauliZ(q)) term = term @ next_var total_hamiltonian += (sign * coeff) * term simplified_H = qml.simplify(total_hamiltonian) coeffs, ops = simplified_H.terms() return qml.Hamiltonian(coeffs, ops)
[docs] def get_mixer_hamiltonian(self, num_wires: int = None) -> qml.Hamiltonian: """ Constructs the Mixer Hamiltonian: Σ X_i. This can be for the full system or a subgroup. Parameters ---------- num_wires : int, optional The number of wires (qubits) for the mixer. If None, uses the total number of qubits. Returns ------- qml.Hamiltonian The mixer Hamiltonian. """ if num_wires is None: num_wires = self.n_qubits return qml.Hamiltonian([1.0] * num_wires, [qml.PauliX(i) for i in range(num_wires)])
[docs] def get_state_vector(self, s: str, weights: List[float], time: float, r: int, mix_weight: float) -> np.ndarray: """ Evolve the initial state and return the final state vector. Parameters ---------- s : str Input bitstring representing the initial state. weights : list of float Coefficients for the problem Hamiltonian terms. time : float Total evolution time. r : int Number of Trotter steps for the approximate time evolution. mix_weight : float Coefficient for the mixer Hamiltonian (between 0 and 1). Returns ------- np.ndarray The final state vector. Notes ----- The total Hamiltonian simulated by the circuit is a weighted sum of the problem Hamiltonian terms and the mixer Hamiltonian: In Ferguson et al. (2025) [arXiv:2506.19538], we use gammas to weight the entire problem Hamiltonian vs the mixer vs the constraint Hamiltonian, such that the total Hamiltonian is: ``H = g_p * H_p + g_m * H_m + g_c * H_c`` but here we allow for separate weights for each coupling tensor term, as well as a separate gamma for the mixer. The total Hamiltonian is then: ``H = (w_b1*H_b1 + w_b2*H_b2 + ... + w_b2*H_bm) + g_m * H_m`` In other words, the constraint hamiltonian is absorbed in the coupling list, and weighted by the corresponding gamma in the gammas list. This allows for more flexible weighting of different terms, and also allows us to use the same code for both constrained and unconstrained problems (by simply including or excluding the constraint Hamiltonian in the coupling list and adjusting the gammas accordingly). Note that it is assumed that each term is already normalised appropriately, so the gammas can be interpreted as the relative weights of each term in the total Hamiltonian. """ num_wires = len(s) dev = self._get_device(num_wires) if mix_weight < 0 or mix_weight > 1: raise ValueError(f"mix_weight must be between 0 and 1. Got {mix_weight}") if np.any(np.array(weights) < 0): raise ValueError(f"Weights must be non-negative. Got {weights}") coeff_mixer = mix_weight coeff_problem = weights H_total = qml.Hamiltonian( [coeff_mixer] + list(np.ones(len(self.model.normalised_couplings))), [self.get_mixer_hamiltonian(num_wires)] + [self.get_problem_hamiltonian(couplings=[self.model.normalised_couplings[i]], sign=coeff_problem[i]) for i in range(len(self.model.normalised_couplings))], ) @qml.qnode(dev) def quantum_evolution(input_string): for i, bit in enumerate(input_string): if bit == "1": qml.PauliX(i) qml.ApproxTimeEvolution(H_total, time, r) return qml.state() state_vector = quantum_evolution(s) return state_vector
[docs] def get_sample(self, s_cg: str, time: float, r: int, mix_weight: float, local_couplings: list, weights: List[float] = None) -> str: """ Generate a single sample by evolving the system and measuring. Parameters ---------- s_cg : str Input bitstring for the subgroup. time : float Total evolution time. r : int Number of Trotter steps for the approximate time evolution. mix_weight : float Coefficient for the mixer Hamiltonian. local_couplings : list Coupling tensors for the subgroup. weights : list of float, optional Coefficients for the problem Hamiltonian terms. Defaults to ones. Returns ------- str A single bitstring sample. """ if weights is None: weights = np.ones(len(local_couplings)) num_wires = len(s_cg) dev = self._get_device(num_wires) if mix_weight < 0 or mix_weight > 1: raise ValueError(f"mix_weight must be between 0 and 1. Got {mix_weight}") if np.any(np.array(weights) < 0): raise ValueError(f"Weights must be non-negative. Got {weights}") coeff_mixer = mix_weight coeff_problem = weights # coeff_problem = [coeff for i, coeff in enumerate(coeff_problem) if local_couplings[i].ndim > 0 and np.any(local_couplings[i] != 0)] # local_couplings = [coupling for coupling in local_couplings if coupling.ndim > 0 and np.any(coupling != 0)] if len(local_couplings) == 0: # If there are no local couplings, just return the input state print("no non-zero local couplings, skipping evolution and returning input state") return s_cg H_total = qml.Hamiltonian( [coeff_mixer] + list(np.ones(len(local_couplings))), [self.get_mixer_hamiltonian(num_wires)] + [self.get_problem_hamiltonian(couplings=[local_couplings[i]], sign=coeff_problem[i]) for i in range(len(local_couplings))], ) # set qnode to use our device with dynamically chosen wires @qml.qnode(dev, shots=1) def quantum_evolution(input_string): for i, bit in enumerate(input_string): if bit == "1": qml.PauliX(i) qml.ApproxTimeEvolution(H_total, time, r) return qml.sample() """print(f"Simulating quantum circuit with {num_wires} qubits, time={time}, num_trotter_steps={num_trotter_steps}") compiled_circuit = qml.compile(quantum_evolution) specs_dict = qml.specs(compiled_circuit)(s_cg) specs_dict = specs_dict['resources'] # 3. Print the relevant metrics print(f"Circuit Depth: {specs_dict['depth']}") print(f"Total Gates: {specs_dict['num_gates']}") # To see the count of two-qubit gates specifically: two_qubit_gates = specs_dict['gate_sizes'].get(2, 0) print(f"Two-qubit Gates: {two_qubit_gates}") # To see the breakdown by gate type (e.g., CNOT, RZ, etc.) print("\nGate Breakdown:") for gate, count in specs_dict['gate_types'].items(): print(f"- {gate}: {count}")""" # Get the first shot from the sample compiled_circuit = qml.compile(quantum_evolution) sample = compiled_circuit(s_cg)[0] # sample = quantum_evolution(s_cg)[0] bitstring = "".join(str(int(b)) for b in sample) return bitstring
[docs] def update(self, s: str, subgroup_choice: List[int], local_couplings: list, gamma: float, time: float, r: int) -> str: """ Update a bitstring by evolving a subgroup. This performs a time evolution on a coarse-grained Hamiltonian to get s' from s. Parameters ---------- s : str The initial bitstring. subgroup_choice : list of int Indices of the subgroup to evolve. local_couplings : list Coupling tensors for the subgroup. gamma : float Mixing parameter. time : float Evolution time. r : int Number of Trotter steps for the approximate time evolution. Returns ------- str The updated bitstring s'. """ self._validate_bitstring(s) # Get s_cg' for the subgroup and reconstruct full s' using s and s_cg' s_cg = "".join([s[i] for i in subgroup_choice]) s_cg_prime = self.get_sample(s_cg, time, r, gamma, local_couplings) s_list = list(s) for i, global_index in enumerate(subgroup_choice): s_list[global_index] = s_cg_prime[i] return "".join(s_list)
[docs] def _validate_bitstring(self, s: str, *, length: int = None): """ Validate s is a bitstring. Parameters ---------- s : str The string to validate. length : int, optional The expected length of the string. Raises ------ TypeError If s is not a string. ValueError If s contains characters other than '0' or '1', or if its length is wrong. """ if not isinstance(s, str): raise TypeError(f"bitstring must be of type str, got {type(s)}: {s!r}") if length is not None and len(s) != length: raise ValueError(f"bitstring must have length {length}, got {len(s)}: {s!r}") bad = set(s) - {"0", "1"} if bad: raise ValueError(f"bitstring must contain only '0'/'1'. Bad chars: {bad}. Value: {s!r}") return s