Trajectree 0.0.4__py3-none-any.whl → 0.0.5__py3-none-any.whl

trajectree/quant_info/circuit.py

@@ -0,0 +1,251 @@
+from quimb.tensor import MatrixProductOperator as mpo
+from trajectree.trajectory import quantum_channel, trajectory_evaluator
+import numpy as np
+from qutip_qip.operations import H, CNOT, S, T, X, Z, CZ, RY, RZ
+from .noise_models import amplitude_damping, phase_damping
+from trajectree.fock_optics.utils import create_vacuum_state
+from qutip import basis, tensor
+import time
+from scipy import sparse as sp
+import copy
+from functools import lru_cache
+
+
+
+class Circuit:
+    def __init__(self, num_qubits=-1, backend = 'tensor'):
+        self.num_qubits = num_qubits
+        self.quantum_channel_list = []
+        self.expectation_ops = []
+        self.backend = backend
+
+        if num_qubits > 0:
+            trajectree_init = [sp.eye(2)]
+            self.quantum_channel_list.append(quantum_channel(N = 2, num_modes = num_qubits, formalism = "kraus", kraus_ops_tuple = ((0,), trajectree_init), backend = self.backend, name = "trajectree_init"))
+            if self.backend ==  'tensor':
+                self.psi = create_vacuum_state(num_qubits, N=2)
+            elif self.backend == 'statevector':
+                self.psi = tensor([basis(2, 0)] * self.num_qubits)
+
+    def create_trajectree(self, cache_size=2, max_cache_nodes=-1):
+        self.t_eval = trajectory_evaluator(self.quantum_channel_list, cache_size = cache_size, max_cache_nodes = max_cache_nodes, backend = self.backend, calc_expectation = True)
+
+    def H_gate(self, idx, is_expectation = False, tag = "H gate"):
+        if self.backend == 'statevector':
+            quantum_channel_H = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], H(0).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == "tensor":
+            dense_op = H(0).get_compact_qobj().full()
+            H_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_H = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = H_MPO, backend = self.backend, name = tag)
+        
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_H)
+        else:
+            self.quantum_channel_list.append(quantum_channel_H)
+
+    def CNOT_gate(self, control_idx, target_idx, tag = "CNOT gate"):
+        if self.backend == 'statevector':
+            quantum_channel_CX = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([control_idx, target_idx], CNOT(0,1).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = CNOT(0,1).get_compact_qobj().full()
+            CNOT_MPO = mpo.from_dense(dense_op, dims = 2, sites = (control_idx, target_idx), L=self.num_qubits, tags=tag)
+            quantum_channel_CX = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = CNOT_MPO, backend = self.backend, name = tag)
+
+        self.quantum_channel_list.append(quantum_channel_CX)
+
+    
+    def CZ_gate(self, control_idx, target_idx, tag = "CZ gate"):
+        if self.backend == 'statevector':
+            quantum_channel_CZ = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([control_idx, target_idx], CZ(0,1).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = CZ(0,1).get_compact_qobj().full()
+            CZ_MPO = mpo.from_dense(dense_op, dims = 2, sites = (control_idx, target_idx), L=self.num_qubits, tags=tag)
+            quantum_channel_CZ = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = CZ_MPO, backend = self.backend, name = tag)
+
+        self.quantum_channel_list.append(quantum_channel_CZ)
+
+
+    def S_gate(self, idx, is_expectation = False, tag = "S gate"):
+        if self.backend == 'statevector':
+            quantum_channel_S = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], S(0).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = S(0).get_compact_qobj().full()
+            S_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_S = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = S_MPO, backend = self.backend, name = tag)
+
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_S)
+        else:
+            self.quantum_channel_list.append(quantum_channel_S)
+
+    
+    def RY_gate(self, theta, idx, is_expectation = False, tag = "RY gate"):
+        if self.backend == 'statevector':
+            quantum_channel_RY = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], RY(0, theta).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = RY(0, theta).get_compact_qobj().full()
+            RY_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_RY = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = RY_MPO, backend = self.backend, name = tag)
+
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_RY)
+        else:
+            self.quantum_channel_list.append(quantum_channel_RY)
+
+    def RZ_gate(self, theta, idx, is_expectation = False, tag = "RZ gate"):
+        if self.backend == 'statevector':
+            quantum_channel_RZ = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], RZ(0, theta).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = RZ(0, theta).get_compact_qobj().full()
+            RZ_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_RZ = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = RZ_MPO, backend = self.backend, name = tag)
+
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_RZ)
+        else:
+            self.quantum_channel_list.append(quantum_channel_RZ)
+
+    def T_gate(self, idx, is_expectation = False, tag = "T gate"):
+        if self.backend == 'statevector':
+            quantum_channel_T = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], T(0).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = T(0).get_compact_qobj().full()
+            T_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_T = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = T_MPO, backend = self.backend, name = tag)
+        
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_T)
+        else:
+            self.quantum_channel_list.append(quantum_channel_T)
+
+
+    def X_gate(self, idx, is_expectation = False, tag = "X gate"):
+        if self.backend == "statevector":
+            quantum_channel_X = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], X(0).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == "tensor":
+            dense_op = X(0).get_compact_qobj().full()
+            X_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_X = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = X_MPO, backend = self.backend, name = tag)
+        
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_X)
+        else:
+            self.quantum_channel_list.append(quantum_channel_X)
+
+
+    def Z_gate(self, idx, is_expectation = False, tag = "X gate"):
+        if self.backend == 'statevector':
+            quantum_channel_Z = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = ([idx], Z(0).get_compact_qobj()), backend = self.backend, name = tag)
+        elif self.backend == 'tensor':
+            dense_op = Z(0).get_compact_qobj().full()
+            Z_MPO = mpo.from_dense(dense_op, dims = 2, sites = (idx,), L=self.num_qubits, tags=tag)
+            quantum_channel_Z = quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "closed", unitary_op = Z_MPO, backend = self.backend, name = tag)
+        
+        if is_expectation:
+            self.expectation_ops.append(quantum_channel_Z)
+        else:
+            self.quantum_channel_list.append(quantum_channel_Z)
+
+    def amplitude_damping(self, noise_parameter, idx, tag = "amplitude_damping"):
+        qc = Circuit.generate_amplitude_damping_quantum_channel(self.num_qubits, self.backend, noise_parameter, idx, tag = tag)
+        self.quantum_channel_list.append(qc)
+
+    @lru_cache(maxsize=100)
+    @staticmethod
+    def generate_amplitude_damping_quantum_channel(num_qubits, backend, noise_parameter, idx, tag = "amplitude_damping"):
+        damping_channels = amplitude_damping(noise_parameter = noise_parameter)
+        return quantum_channel(N = 2, num_modes = num_qubits, formalism = "kraus", kraus_ops_tuple = ((idx,), damping_channels), backend = backend, name = tag)
+        
+
+    def phase_damping(self, noise_parameter, idx, tag = "phase_damping"):
+        damping_channels = phase_damping(noise_parameter = noise_parameter)
+        self.quantum_channel_list.append(quantum_channel(N = 2, num_modes = self.num_qubits, formalism = "kraus", kraus_ops_tuple = ((idx,), damping_channels), backend = self.backend, name = tag))
+    
+    def perform_trajectree_simulation(self, num_simulations, error_tolerance = 1e-12, verbose = False):
+        times = []
+        evs = []
+        for _ in range(num_simulations): 
+            start = time.time()
+            evs.append(self.t_eval.perform_simulation(self.psi, error_tolerance, normalize = True))
+        
+            time_taken = time.time() - start
+            # print("time taken:", time_taken)
+            if verbose:
+                print("time taken:", time_taken)
+            #     if i in progress:
+            #         print(f"Completed {progress.index(i)+1}0% of simulations")
+            times.append(time_taken)
+        # print("done with simulations")
+        
+        return evs, times
+
+    def qiskit_to_trajectree(self, qc, noise_parameter):
+        self.__init__(qc.num_qubits, backend = self.backend) 
+
+        for circuit_instr in qc.data:
+            instr = circuit_instr.operation
+            qargs = circuit_instr.qubits
+            cargs = circuit_instr.clbits
+            gate_name = instr.name
+            
+            if gate_name == 'x':
+                self.X_gate(qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 'h':
+                self.H_gate(qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 's':
+                self.S_gate(qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 't':
+                self.T_gate(qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 'z':
+                self.Z_gate(qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 'cx':
+                self.CNOT_gate(qargs[0]._index, qargs[1]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[1]._index)
+
+            elif gate_name == 'cz':
+                self.CZ_gate(qargs[0]._index, qargs[1]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[1]._index)
+
+            elif gate_name == 'ry':
+                self.RY_gate(instr.params[0], qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+            elif gate_name == 'rz':
+                self.RZ_gate(instr.params[0], qargs[0]._index)
+                self.amplitude_damping(noise_parameter=noise_parameter, idx = qargs[0]._index)
+
+        self.create_trajectree()
+
+    def expectation(self, ops):
+        for i in range(len(ops)):
+            if ops[i] == 'x':
+                self.X_gate(i, is_expectation = True)
+
+            elif ops[i] == 'h':
+                self.H_gate(i, is_expectation = True)
+
+            elif ops[i] == 's':
+                self.S_gate(i, is_expectation = True)
+
+            elif ops[i] == 't':
+                self.T_gate(i, is_expectation = True)
+
+            elif ops[i] == 'z':
+                self.Z_gate(i, is_expectation = True)
+            else:
+                continue
+        self.t_eval.observable_ops = self.expectation_ops
+        self.t_eval.calc_expectation = True
+

trajectree/quant_info/noise_models.py

@@ -0,0 +1,24 @@
+from scipy import sparse as sp
+from scipy.linalg import expm
+
+import numpy as np
+from numpy import sqrt
+import qutip as qt
+from math import factorial
+from functools import lru_cache
+
+@lru_cache(maxsize=100)
+def amplitude_damping(noise_parameter):
+
+    ops = []
+    ops.append(sp.csr_array([[1, 0], [0, sqrt(1 - noise_parameter)]]))
+    ops.append(sp.csr_array([[0, sqrt(noise_parameter)], [0, 0]]))
+    return ops
+
+@lru_cache(maxsize=100)
+def phase_damping(noise_parameter):
+
+    ops = []
+    ops.append(sp.csr_array([[1, 0], [0, sqrt(1 - noise_parameter)]]))
+    ops.append(sp.csr_array([[0, 0], [0, sqrt(noise_parameter)]]))
+    return ops  

trajectree/sequence/swap.py

@@ -10,28 +10,33 @@ import time
 import numpy as np
 import copy
 
-def generate_swapping_circuit(N, num_modes, site_tags, bsm_det_effs, bsm_dark_counts_gain, bsm_measurements,  channel_loss, error_tolerance):
+def generate_swapping_circuit(N, num_modes, site_tags, bsm_det_effs, bsm_dark_counts_gain, bsm_measurements, channel_loss, depolarizing_error = False, damping_error = True,  error_tolerance=1e-10):
     quantum_channel_list = []
 
     # Amplitude damping due to fibers
-    damping_kraus_ops = single_mode_bosonic_noise_channels(noise_parameter = channel_loss, N = N)
-    two_mode_kraus_ops = [sp.kron(op, op) for op in damping_kraus_ops]
-    quantum_channel_list.append(quantum_channel(N = N, num_modes = num_modes, formalism = "kraus", kraus_ops_tuple = ((2,3), two_mode_kraus_ops), name = "fiber_attenuation")) # The tuples in this list are defined as (sites, kraus_ops). The sites are the sites where the Kraus ops are applied.
-    quantum_channel_list.append(quantum_channel(N = N, num_modes = num_modes, formalism = "kraus", kraus_ops_tuple = ((6,7), two_mode_kraus_ops), name = "fiber_attenuation")) # The tuples in this list are defined as (sites, kraus_ops). The sites are the sites where the Kraus ops are applied.
+
+    phase_flip_ops = [sp.csr_matrix(global_phase(0, N, return_unitary = True, site1 = 0, site2 = 1, total_sites = 8))]
+    quantum_channel_list.append(quantum_channel(N = N, num_modes = 8, formalism = "kraus", kraus_ops_tuple = ((0,1), phase_flip_ops), name = "first_kraus"))
+
+
+    # damping_kraus_ops = single_mode_bosonic_noise_channels(noise_parameter = 1-1e-3, N = N)
+    # two_mode_kraus_ops = [[sp.kron(op1, op2) for op1 in damping_kraus_ops for op2 in damping_kraus_ops][0]]
+    # quantum_channel_list.append(quantum_channel(N = N, num_modes = num_modes, formalism = "kraus", kraus_ops_tuple = ((2,3), two_mode_kraus_ops), name = "fiber_attenuation")) # The tuples in this list are defined as (sites, kraus_ops). The sites are the sites where the Kraus ops are applied.
+    # quantum_channel_list.append(quantum_channel(N = N, num_modes = num_modes, formalism = "kraus", kraus_ops_tuple = ((6,7), two_mode_kraus_ops), name = "fiber_attenuation")) # The tuples in this list are defined as (sites, kraus_ops). The sites are the sites where the Kraus ops are applied.
 
     # Quantum channel for the Bell state measurement
     # BSM_MPOs = bell_state_measurement(None, N, site_tags, num_modes, bsm_det_effs, error_tolerance, measurements = bsm_measurements, pnr = False, use_trajectory = True, return_MPOs = True, compress=True, contract=True)
     # BSM_quantum_channels = [quantum_channel(N = N, num_modes = num_modes, formalism = "closed", unitary_MPOs = BSM_MPO, name = "BSM") for BSM_MPO in BSM_MPOs]
-    BSM_quantum_channels = bell_state_measurement(None, N, site_tags, num_modes, bsm_det_effs, bsm_dark_counts_gain, error_tolerance, measurements = bsm_measurements, pnr = False, use_trajectory = True, return_MPOs = True, compress=True, contract=True)
+    
+    BSM_quantum_channels, expectation_channels = bell_state_measurement(None, N, site_tags, num_modes, bsm_det_effs, bsm_dark_counts_gain, error_tolerance, measurements = bsm_measurements, depolarizing_error = depolarizing_error, damping_error = damping_error, pnr = False, use_trajectory = True, return_MPOs = True, compress=True, contract=True)
     quantum_channel_list.extend(BSM_quantum_channels)
 
-    return quantum_channel_list
+    return quantum_channel_list, expectation_channels
 
-def analyze_entanglement(quantum_channel_list, N, site_tags, num_modes, efficiency, error_tolerance, idler_angles, signal_angles):
-    PA_MPOs = rotate_and_measure(None, N, site_tags, num_modes, efficiency, error_tolerance, idler_angles, signal_angles, return_MPOs = True)
-    PA_quantum_channels = [quantum_channel(N = N, num_modes = num_modes, formalism = "closed", unitary_MPOs = PA_MPO) for PA_MPO in PA_MPOs]
-    # print("num pa quantum channels:", len(PA_quantum_channels))
+def analyze_entanglement(quantum_channel_list, expectation_channels, N, site_tags, num_modes, efficiency, error_tolerance, idler_angles, signal_angles, depolarizing_error = False, damping_error = True):
+    PA_quantum_channels, PA_expectation_channels = rotate_and_measure(None, N, site_tags, num_modes, efficiency, error_tolerance, idler_angles, signal_angles, depolarizing_error = depolarizing_error, damping_error = damping_error, return_quantum_channel = True)
     quantum_channel_list.extend(PA_quantum_channels)
+    expectation_channels.extend(PA_expectation_channels)
 
 
 def create_swapping_initial_state(num_modes, N, mean_photon_num, error_tolerance):
@@ -44,34 +49,53 @@ def create_swapping_initial_state(num_modes, N, mean_photon_num, error_tolerance
     psi = extend_MPS(psi)
     return psi
 
-def perform_swapping_simulation(N, num_modes, num_simulations, params, error_tolerance = 1e-10):
-
+def create_swapping_simulation(N, num_modes, params, cache_size, error_tolerance = 1e-10):
     psi = create_swapping_initial_state(num_modes, N, params["chi"], error_tolerance)
 
-    quantum_channels = generate_swapping_circuit(N, num_modes, psi.site_tags, [params["BSM_det_loss_1"], params["BSM_det_loss_2"]], [params["BSM_dark_counts_1"], params["BSM_dark_counts_2"]], params["BSM_meas"], params["channel_loss"], error_tolerance)
+    quantum_channels, expectation_channels = generate_swapping_circuit(N, num_modes, psi.site_tags, [params["BSM_det_loss_1"], params["BSM_det_loss_2"]], [params["BSM_dark_counts_1"], params["BSM_dark_counts_2"]], params["BSM_meas"], params["channel_loss"], depolarizing_error = params["depolarizing_error"], damping_error = params["damping_error"], error_tolerance = error_tolerance)
 
     if params["if_analyze_entanglement"]:
-        analyze_entanglement(quantum_channels, N, psi.site_tags, num_modes, params["PA_det_eff"], error_tolerance, params["alpha_list"], params["delta_list"])
+        analyze_entanglement(quantum_channels, expectation_channels, N, psi.site_tags, num_modes, params["PA_det_eff"], error_tolerance, params["alpha_list"], params["delta_list"], depolarizing_error = params["depolarizing_error"], damping_error = params["damping_error"])
+
+    # quantum_channels.extend(expectation_channels)
 
-    t_eval = trajectory_evaluator(quantum_channels)
+    t_eval = trajectory_evaluator(quantum_channels, cache_size = cache_size, max_cache_nodes = params["max_cache_nodes"], calc_expectation = True, observable_ops = expectation_channels)
+
+    return psi, t_eval
+
+
+def perform_swapping_simulation(psi, t_eval, num_simulations, verbose = False, N = None, calc_fidelity = False, error_tolerance = 1e-10):
+    # print("initial state:")
+    # read_quantum_state(psi, N)
+
+    # phase_flip_ops = [sp.csr_matrix(global_phase(0, N, return_unitary = True, site1 = 0, site2 = 1, total_sites = 2))]
+    # quantum_channels.append(quantum_channel(N = N, num_modes = 8, formalism = "kraus", kraus_ops_tuple = ((0,1), phase_flip_ops), name = "first_kraus"))
 
     fidelities = []
     probabilities = []
+    times = []
+    progress = [x * num_simulations//10 for x in range(1, 11)]
 
     for i in range(num_simulations): 
         start = time.time()
-        psi_iter = copy.deepcopy(t_eval.perform_simulation(psi, error_tolerance, normalize = False))
+        psi_iter = probabilities.append(t_eval.perform_simulation(psi, error_tolerance, normalize = True))
 
-        probabilities.append(psi_iter.normalize())
+        # # read_quantum_state(psi_iter, N)
+        # if psi_iter == 0:
+        #     probabilities.append(0)
+        # else:
+        #     probabilities.append(psi_iter.normalize())
         
-        if params["calc_fidelity"]:
-            fidelity = np.abs(calc_fidelity_swapping(psi_iter, "psi_plus", N, error_tolerance))
-            fidelities.append(fidelity)
+        # if calc_fidelity:
+        #     fidelity = np.abs(calc_fidelity_swapping(psi_iter, "psi_minus", N, error_tolerance))
+        #     fidelities.append(fidelity)
     
         time_taken = time.time() - start
-        # print("time taken for simulation", i, ":", time_taken)
+        if verbose:
+            if i in progress:
+                print(f"Completed {progress.index(i)+1}0% of simulations")
+        times.append(time_taken)
 
     # print("completed set", "cache_hits:", t_eval.cache_hit, "cache_partial_hits:", t_eval.cache_partial_hit, "cache_misses:", t_eval.cache_miss,  "time taken:", time_taken)
 
-    return fidelities, probabilities, t_eval
-
+    return fidelities, probabilities, times