Make QGAN run primarily on circuits (#1341)

* make quantum generator use circuits primarily

* add test and deprecation warning

* add reno

* fix lint

Co-authored-by: Manoel Marques <[email protected]>

Author: Cryoris

qiskit/aqua/components/neural_networks/generative_network.py

@@ -43,13 +43,12 @@ def set_seed(self, seed):
         raise NotImplementedError()
 
     @abstractmethod
-    def get_output(self, quantum_instance, qc_state_in, params, shots):
+    def get_output(self, quantum_instance, params, shots):
         """
         Apply quantum/classical neural network to given input and get the respective output
 
         Args:
             quantum_instance (QuantumInstance): Quantum Instance, used to run the generator circuit.
-            qc_state_in (QuantumCircuit or vector): corresponding to the network input state
             params (numpy.ndarray): parameters which should be used to run the generator,
                 if None use self._params
             shots (int): if not None use a number of shots that is different from the number

qiskit/aqua/components/neural_networks/quantum_generator.py

@@ -13,20 +13,17 @@
 """Quantum Generator."""
 
 from typing import Optional, List, Union, Dict, Any
+import warnings
 from copy import deepcopy
 import numpy as np
 
 from qiskit import QuantumRegister, ClassicalRegister, QuantumCircuit
 from qiskit.circuit.library import TwoLocal
 from qiskit.aqua import aqua_globals
 from qiskit.aqua.components.optimizers import ADAM
-from qiskit.aqua.components.uncertainty_models import \
-    UniformDistribution, MultivariateUniformDistribution
 from qiskit.aqua.components.uncertainty_models import UnivariateVariationalDistribution, \
     MultivariateVariationalDistribution
-from qiskit.aqua import AquaError
 from qiskit.aqua.components.neural_networks.generative_network import GenerativeNetwork
-from qiskit.aqua.components.initial_states import Custom
 
 # pylint: disable=invalid-name
 
@@ -72,61 +69,32 @@ def __init__(self,
         self._bounds = bounds
         self._num_qubits = num_qubits
         self.generator_circuit = generator_circuit
-        if self.generator_circuit is None:
-            entangler_map = []
-            if np.sum(num_qubits) > 2:
-                for i in range(int(np.sum(num_qubits))):
-                    entangler_map.append([i, int(np.mod(i + 1, np.sum(num_qubits)))])
-            else:
-                if np.sum(num_qubits) > 1:
-                    entangler_map.append([0, 1])
-
-            if len(num_qubits) > 1:
-                num_qubits = list(map(int, num_qubits))
-                low = bounds[:, 0].tolist()
-                high = bounds[:, 1].tolist()
-                init_dist = MultivariateUniformDistribution(num_qubits, low=low, high=high)
-                q = QuantumRegister(sum(num_qubits))
-                qc = QuantumCircuit(q)
-                init_dist.build(qc, q)
-                init_distribution = Custom(num_qubits=sum(num_qubits), circuit=qc)
-                # Set variational form
-                var_form = TwoLocal(sum(num_qubits), 'ry', 'cz', reps=1,
-                                    initial_state=init_distribution,
-                                    entanglement=entangler_map)
-                if init_params is None:
-                    init_params = aqua_globals.random.random(var_form.num_parameters) * 2 * 1e-2
-                # Set generator circuit
-                self.generator_circuit = MultivariateVariationalDistribution(num_qubits, var_form,
-                                                                             init_params,
-                                                                             low=low, high=high)
-            else:
-                init_dist = UniformDistribution(sum(num_qubits), low=bounds[0], high=bounds[1])
-                q = QuantumRegister(sum(num_qubits), name='q')
-                qc = QuantumCircuit(q)
-                init_dist.build(qc, q)
-                init_distribution = Custom(num_qubits=sum(num_qubits), circuit=qc)
-                var_form = TwoLocal(sum(num_qubits), 'ry', 'cz', reps=1,
-                                    initial_state=init_distribution,
-                                    entanglement=entangler_map)
-                if init_params is None:
-                    init_params = aqua_globals.random.random(var_form.num_parameters) * 2 * 1e-2
-                # Set generator circuit
-                self.generator_circuit = UnivariateVariationalDistribution(
-                    int(np.sum(num_qubits)), var_form, init_params, low=bounds[0], high=bounds[1])
-
-        if len(num_qubits) > 1:
-            if isinstance(self.generator_circuit, MultivariateVariationalDistribution):
-                pass
-            else:
-                raise AquaError('Set multivariate variational distribution '
-                                'to represent multivariate data')
+        if generator_circuit is None:
+            circuit = QuantumCircuit(sum(num_qubits))
+            circuit.h(circuit.qubits)
+            var_form = TwoLocal(sum(num_qubits), 'ry', 'cz', reps=1, entanglement='circular')
+            circuit.compose(var_form, inplace=True)
+
+            # Set generator circuit
+            self.generator_circuit = circuit
+
+        if isinstance(generator_circuit, (UnivariateVariationalDistribution,
+                                          MultivariateVariationalDistribution)):
+            warnings.warn('Passing a UnivariateVariationalDistribution or MultivariateVariational'
+                          'Distribution is as ``generator_circuit`` is deprecated as of Aqua 0.8.0 '
+                          'and the support will be removed no earlier than 3 months after the '
+                          'release data. You should pass as QuantumCircuit instead.',
+                          DeprecationWarning, stacklevel=2)
+            self._free_parameters = generator_circuit._var_form_params
+            self.generator_circuit = generator_circuit._var_form
         else:
-            if isinstance(self.generator_circuit, UnivariateVariationalDistribution):
-                pass
-            else:
-                raise AquaError('Set univariate variational distribution '
-                                'to represent univariate data')
+            self._free_parameters = list(self.generator_circuit.parameters)
+
+        if init_params is None:
+            init_params = aqua_globals.random.random(self.generator_circuit.num_parameters) * 2e-2
+
+        self._bound_parameters = init_params
+
         # Set optimizer for updating the generator network
         self._optimizer = ADAM(maxiter=1, tol=1e-6, lr=1e-3, beta_1=0.7,
                                beta_2=0.99, noise_factor=1e-6,
@@ -192,26 +160,28 @@ def construct_circuit(self, params=None):
         Construct generator circuit.
 
         Args:
-            params (numpy.ndarray): parameters which should be used to run the generator,
-                    if None use self._params
+            params (list | dict): parameters which should be used to run the generator.
 
         Returns:
             Instruction: construct the quantum circuit and return as gate
         """
-
-        q = QuantumRegister(sum(self._num_qubits), name='q')
-        qc = QuantumCircuit(q)
         if params is None:
-            self.generator_circuit.build(qc=qc, q=q)
-        else:
-            generator_circuit_copy = deepcopy(self.generator_circuit)
-            generator_circuit_copy.params = params
-            generator_circuit_copy.build(qc=qc, q=q)
+            return self.generator_circuit
 
-        # return qc.copy(name='qc')
-        return qc.to_instruction()
+        if isinstance(params, (list, np.ndarray)):
+            params = dict(zip(self._free_parameters, params))
 
-    def get_output(self, quantum_instance, qc_state_in=None, params=None, shots=None):
+        return self.generator_circuit.assign_parameters(params)
+        #     self.generator_circuit.build(qc=qc, q=q)
+        # else:
+        #     generator_circuit_copy = deepcopy(self.generator_circuit)
+        #     generator_circuit_copy.params = params
+        #     generator_circuit_copy.build(qc=qc, q=q)
+
+        # # return qc.copy(name='qc')
+        # return qc.to_instruction()
+
+    def get_output(self, quantum_instance, params=None, shots=None):
         """
         Get classical data samples from the generator.
         Running the quantum generator circuit results in a quantum state.
@@ -222,7 +192,6 @@ def get_output(self, quantum_instance, qc_state_in=None, params=None, shots=None
         Args:
             quantum_instance (QuantumInstance): Quantum Instance, used to run the generator
                 circuit.
-            qc_state_in (QuantumCircuit): deprecated
             params (numpy.ndarray): array or None, parameters which should
                 be used to run the generator, if None use self._params
             shots (int): if not None use a number of shots that is different from the
@@ -234,6 +203,8 @@ def get_output(self, quantum_instance, qc_state_in=None, params=None, shots=None
         instance_shots = quantum_instance.run_config.shots
         q = QuantumRegister(sum(self._num_qubits), name='q')
         qc = QuantumCircuit(q)
+        if params is None:
+            params = self._bound_parameters
         qc.append(self.construct_circuit(params), q)
         if quantum_instance.is_statevector:
             pass
@@ -277,7 +248,7 @@ def get_output(self, quantum_instance, qc_state_in=None, params=None, shots=None
                     temp.append(self._data_grid[int(bin_rep)])
             generated_samples.append(temp)
 
-        self.generator_circuit._probabilities = generated_samples_weights
+        # self.generator_circuit._probabilities = generated_samples_weights
         if shots is not None:
             # Restore the initial quantum_instance configuration
             quantum_instance.set_config(shots=instance_shots)
@@ -347,13 +318,13 @@ def train(self, quantum_instance=None, shots=None):
         self._optimizer._maxiter = 1
         self._optimizer._t = 0
         objective = self._get_objective_function(quantum_instance, self._discriminator)
-        self.generator_circuit.params, loss, _ = self._optimizer.optimize(
-            num_vars=len(self.generator_circuit.params),
+        self._bound_parameters, loss, _ = self._optimizer.optimize(
+            num_vars=len(self._bound_parameters),
             objective_function=objective,
-            initial_point=self.generator_circuit.params
+            initial_point=self._bound_parameters
             )
 
         self._ret['loss'] = loss
-        self._ret['params'] = self.generator_circuit.params
+        self._ret['params'] = self._bound_parameters
 
         return self._ret

releasenotes/notes/qgan-on-circuits-9bd57bd707b31897.yaml

@@ -0,0 +1,11 @@
+---
+features:
+  - |
+    Support passing ``QuantumCircuit`` objects as generator circuits into 
+    the ``QuantumGenerator``.
+deprecations:
+  - |
+    Deprecate the ``UnivariateVariationalDistribution`` and 
+    ``MultivariateVariationalDistribution`` as input 
+    to the ``QuantumGenerator``. Instead, plain ``QuantumCircuit`` objects can 
+    be used.

test/aqua/test_qgan.py

@@ -12,20 +12,23 @@
 
 """Test the QGAN algorithm."""
 
+import unittest
+import warnings
 from test.aqua import QiskitAquaTestCase
+from ddt import ddt, data
 
-import unittest
 from qiskit import QuantumCircuit, QuantumRegister
 from qiskit.circuit.library import RealAmplitudes
 from qiskit.aqua.components.uncertainty_models import (UniformDistribution,
                                                        UnivariateVariationalDistribution)
 from qiskit.aqua.algorithms import QGAN
-from qiskit.aqua import aqua_globals, QuantumInstance
+from qiskit.aqua import aqua_globals, QuantumInstance, MissingOptionalLibraryError
 from qiskit.aqua.components.initial_states import Custom
 from qiskit.aqua.components.neural_networks import NumPyDiscriminator, PyTorchDiscriminator
 from qiskit import BasicAer
 
 
+@ddt
 class TestQGAN(QiskitAquaTestCase):
     """Test the QGAN algorithm."""
 
@@ -83,14 +86,24 @@ def setUp(self):
         # Set variational form
         var_form = RealAmplitudes(sum(num_qubits), reps=1, initial_state=init_distribution,
                                   entanglement=entangler_map)
-        self.generator_circuit = UnivariateVariationalDistribution(sum(num_qubits), var_form,
+        self.generator_circuit = var_form
+        warnings.filterwarnings('ignore', category=DeprecationWarning)
+        self.generator_factory = UnivariateVariationalDistribution(sum(num_qubits), var_form,
                                                                    init_params,
                                                                    low=self._bounds[0],
                                                                    high=self._bounds[1])
+        warnings.filterwarnings('always', category=DeprecationWarning)
 
-    def test_sample_generation(self):
+    @data('circuit', 'factory')
+    def test_sample_generation(self, circuit_type):
         """Test sample generation."""
-        self.qgan.set_generator(generator_circuit=self.generator_circuit)
+        if circuit_type == 'factory':
+            warnings.filterwarnings('ignore', category=DeprecationWarning)
+            self.qgan.set_generator(generator_circuit=self.generator_factory)
+            warnings.filterwarnings('always', category=DeprecationWarning)
+        else:
+            self.qgan.set_generator(generator_circuit=self.generator_circuit)
+
         _, weights_statevector = self.qgan._generator.get_output(self.qi_statevector, shots=100)
         samples_qasm, weights_qasm = self.qgan._generator.get_output(self.qi_qasm, shots=100)
         samples_qasm, weights_qasm = zip(*sorted(zip(samples_qasm, weights_qasm)))
@@ -99,7 +112,10 @@ def test_sample_generation(self):
 
     def test_qgan_training(self):
         """Test QGAN training."""
+        warnings.filterwarnings('ignore', category=DeprecationWarning)
         self.qgan.set_generator(generator_circuit=self.generator_circuit)
+        warnings.filterwarnings('always', category=DeprecationWarning)
+
         trained_statevector = self.qgan.run(self.qi_statevector)
         trained_qasm = self.qgan.run(self.qi_qasm)
         self.assertAlmostEqual(trained_qasm['rel_entr'], trained_statevector['rel_entr'], delta=0.1)
@@ -131,8 +147,8 @@ def test_qgan_training_run_algo_torch(self):
                                                      seed_transpiler=aqua_globals.random_seed))
             self.assertAlmostEqual(trained_qasm['rel_entr'],
                                    trained_statevector['rel_entr'], delta=0.1)
-        except Exception as ex:  # pylint: disable=broad-except
-            self.skipTest(str(ex))
+        except MissingOptionalLibraryError:
+            self.skipTest('pytorch not installed, skipping test')
 
     def test_qgan_training_run_algo_numpy(self):
         """Test QGAN training using a NumPy discriminator."""