[Feature] Setting digital noise in addition to readout noise in Pyq backend functions sample and expectation

qadence/backends/pyqtorch/backend.py

@@ -28,6 +28,7 @@
     flatten,
     invert_endianness,
     scale_primitive_blocks_only,
+    set_noise,
     transpile,
 )
 from qadence.types import BackendName, Endianness, Engine
@@ -38,6 +39,46 @@
 logger = getLogger(__name__)
 
 
+def set_noise_abstract_to_native(circuit: ConvertedCircuit, config: Configuration) -> None:
+    """Set noise in native blocks from the abstract ones with noise.
+
+    Args:
+        circuit (ConvertedCircuit): Input converted circuit.
+    """
+    ops = convert_block(circuit.abstract.block, n_qubits=circuit.native.n_qubits, config=config)
+    circuit.native = pyq.QuantumCircuit(circuit.native.n_qubits, ops, circuit.native.readout_noise)
+
+
+def set_readout_noise(circuit: ConvertedCircuit, noise: NoiseHandler) -> None:
+    """Set readout noise in place in native.
+
+    Args:
+        circuit (ConvertedCircuit):  Input converted circuit.
+        noise (NoiseHandler | None): Noise.
+    """
+    readout = convert_readout_noise(circuit.abstract.n_qubits, noise)
+    if readout:
+        circuit.native.readout_noise = readout
+
+
+def set_block_and_readout_noises(
+    circuit: ConvertedCircuit, noise: NoiseHandler | None, config: Configuration
+) -> None:
+    """Add noise on blocks and readout on circuit.
+
+    We first start by adding noise to the abstract blocks. Then we do a conversion to their
+    native representation. Finally, we add readout.
+
+    Args:
+        circuit (ConvertedCircuit): Input circuit.
+        noise (NoiseHandler | None): Noise to add.
+    """
+    if noise:
+        set_noise(circuit, noise)
+        set_noise_abstract_to_native(circuit, config)
+        set_readout_noise(circuit, noise)
+
+
 @dataclass(frozen=True, eq=True)
 class Backend(BackendInterface):
     """PyQTorch backend."""
@@ -55,13 +96,27 @@ class Backend(BackendInterface):
     logger.debug("Initialised")
 
     def circuit(self, circuit: QuantumCircuit) -> ConvertedCircuit:
+        """Return the converted circuit.
+
+        Note that to get a representation with noise, noise
+        should be passed within the config.
+
+        Args:
+            circuit (QuantumCircuit): Original circuit
+
+        Returns:
+            ConvertedCircuit: ConvertedCircuit instance for backend.
+        """
         passes = self.config.transpilation_passes
         if passes is None:
             passes = default_passes(self.config)
 
         original_circ = circuit
         if len(passes) > 0:
             circuit = transpile(*passes)(circuit)
+        # Setting noise in the circuit.
+        if self.config.noise:
+            set_noise(circuit, self.config.noise)
 
         ops = convert_block(circuit.block, n_qubits=circuit.n_qubits, config=self.config)
         readout_noise = (
@@ -124,9 +179,7 @@ def _batched_expectation(
         noise: NoiseHandler | None = None,
         endianness: Endianness = Endianness.BIG,
     ) -> Tensor:
-        if noise and circuit.native.readout_noise is None:
-            readout = convert_readout_noise(circuit.abstract.n_qubits, noise)
-            circuit.native.readout_noise = readout
+        set_block_and_readout_noises(circuit, noise, self.config)
         state = self.run(
             circuit,
             param_values=param_values,
@@ -164,9 +217,7 @@ def _looped_expectation(
                 "Define your initial state with `batch_size=1`"
             )
 
-        if noise and circuit.native.readout_noise is None:
-            readout = convert_readout_noise(circuit.abstract.n_qubits, noise)
-            circuit.native.readout_noise = readout
+        set_block_and_readout_noises(circuit, noise, self.config)
 
         list_expvals = []
         observables = observable if isinstance(observable, list) else [observable]
@@ -222,9 +273,7 @@ def sample(
         elif state is not None and pyqify_state:
             n_qubits = circuit.abstract.n_qubits
             state = pyqify(state, n_qubits) if pyqify_state else state
-        if noise and circuit.native.readout_noise is None:
-            readout = convert_readout_noise(circuit.abstract.n_qubits, noise)
-            circuit.native.readout_noise = readout
+        set_block_and_readout_noises(circuit, noise, self.config)
         samples: list[Counter] = circuit.native.sample(
             state=state, values=param_values, n_shots=n_shots
         )

qadence/backends/pyqtorch/convert_ops.py

@@ -175,8 +175,23 @@ def fn(x: str | ConcretizedCallable, y: str | ConcretizedCallable) -> Callable:
 
 
 def convert_block(
-    block: AbstractBlock, n_qubits: int = None, config: Configuration = None
+    block: AbstractBlock,
+    n_qubits: int = None,
+    config: Configuration = None,
 ) -> Sequence[Module | Tensor | str | sympy.Expr]:
+    """Convert block to native Pyqtorch representation.
+
+    Args:
+        block (AbstractBlock): Block to convert.
+        n_qubits (int, optional): Number of qubits. Defaults to None.
+        config (Configuration, optional): Backend configuration instance. Defaults to None.
+
+    Raises:
+        NotImplementedError: For non supported blocks.
+
+    Returns:
+        Sequence[Module | Tensor | str | sympy.Expr]: List of native operations.
+    """
     if isinstance(block, (Tensor, str, sympy.Expr)):  # case for hamevo generators
         if isinstance(block, Tensor):
             block = block.permute(1, 2, 0)  # put batch size in the back

qadence/noise/protocols.py

@@ -149,15 +149,19 @@ def list(cls) -> list:
         return list(filter(lambda el: not el.startswith("__"), dir(cls)))
 
     def filter(self, protocol: NoiseEnum | str) -> NoiseHandler | None:
-        is_protocol: list = [p == protocol or isinstance(p, protocol) for p in self.protocol]  # type: ignore[arg-type]
-        return (
-            NoiseHandler(
-                list(compress(self.protocol, is_protocol)),
-                list(compress(self.options, is_protocol)),
+        protocol_matches: list = list()
+        if protocol == NoiseProtocol.READOUT:
+            protocol_matches = [p == protocol for p in self.protocol]
+        else:
+            protocol_matches = [isinstance(p, protocol) for p in self.protocol]  # type: ignore[arg-type]
+
+        # if we have at least a match
+        if True in protocol_matches:
+            return NoiseHandler(
+                list(compress(self.protocol, protocol_matches)),
+                list(compress(self.options, protocol_matches)),
             )
-            if len(is_protocol) > 0
-            else None
-        )
+        return None
 
     def bitflip(self, *args: Any, **kwargs: Any) -> NoiseHandler:
         self.append(NoiseHandler(NoiseProtocol.DIGITAL.BITFLIP, *args, **kwargs))

qadence/transpile/noise.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from qadence.backend import ConvertedCircuit
 from qadence.blocks.abstract import AbstractBlock
 from qadence.circuit import QuantumCircuit
 from qadence.noise.protocols import NoiseHandler
@@ -23,24 +24,29 @@ def _set_noise(
 
 
 def set_noise(
-    circuit: QuantumCircuit | AbstractBlock,
+    circuit: QuantumCircuit | AbstractBlock | ConvertedCircuit,
     noise: NoiseHandler | None,
     target_class: AbstractBlock | None = None,
 ) -> QuantumCircuit | AbstractBlock:
     """
     Parses a `QuantumCircuit` or `CompositeBlock` to add noise to specific gates.
 
+    If `circuit` is a `ConvertedCircuit`, this is done within `circuit.abstract`.
+
     Changes the input in place.
 
     Arguments:
         circuit: the circuit or block to parse.
         noise: the NoiseHandler protocol to change to, or `None` to remove the noise.
         target_class: optional class to selectively add noise to.
     """
-    is_circuit_input = isinstance(circuit, QuantumCircuit)
-
-    input_block: AbstractBlock = circuit.block if is_circuit_input else circuit  # type: ignore
+    if isinstance(circuit, ConvertedCircuit):
+        input_block: AbstractBlock = circuit.abstract.block
+    elif isinstance(circuit, QuantumCircuit):
+        input_block = circuit.block
+    else:
+        input_block = circuit
 
-    output_block = apply_fn_to_blocks(input_block, _set_noise, noise, target_class)
+    apply_fn_to_blocks(input_block, _set_noise, noise, target_class)
 
     return circuit

tests/qadence/test_noise/test_digital_noise.py

@@ -92,6 +92,42 @@ def test_run_digital(noisy_config: NoiseProtocol | list[NoiseProtocol]) -> None:
     assert torch.allclose(noisy_output, native_output)
 
 
+@pytest.mark.parametrize(
+    "noisy_config",
+    [
+        NoiseProtocol.DIGITAL.BITFLIP,
+        [NoiseProtocol.DIGITAL.BITFLIP, NoiseProtocol.DIGITAL.PHASEFLIP],
+    ],
+)
+def test_expectation_digital_noise(noisy_config: NoiseProtocol | list[NoiseProtocol]) -> None:
+    block = kron(H(0), Z(1))
+    circuit = QuantumCircuit(2, block)
+    observable = hamiltonian_factory(circuit.n_qubits, detuning=Z)
+    noise = NoiseHandler(noisy_config, {"error_probability": 0.1})
+    backend = backend_factory(backend=BackendName.PYQTORCH, diff_mode=DiffMode.AD)
+
+    # Construct a quantum model.
+    model = QuantumModel(circuit=circuit, observable=observable)
+    noiseless_expectation = model.expectation(values={})
+
+    (pyqtorch_circ, pyqtorch_obs, embed, params) = backend.convert(circuit, observable)
+    native_noisy_expectation = backend.expectation(
+        pyqtorch_circ, pyqtorch_obs, embed(params, {}), noise=noise
+    )
+    assert not torch.allclose(noiseless_expectation, native_noisy_expectation)
+
+    noisy_model = QuantumModel(circuit=circuit, observable=observable, noise=noise)
+    noisy_model_expectation = noisy_model.expectation(values={})
+    assert torch.allclose(noisy_model_expectation, native_noisy_expectation)
+
+    (pyqtorch_circ, pyqtorch_obs, embed, params) = backend.convert(circuit, observable)
+    noisy_converted_model_expectation = backend.expectation(
+        pyqtorch_circ, pyqtorch_obs, embed(params, {})
+    )
+
+    assert torch.allclose(noisy_converted_model_expectation, native_noisy_expectation)
+
+
 @pytest.mark.parametrize(
     "noise_config",
     [