Serialization

Qadence offers convenience functions for serializing and deserializing any quantum program. This is useful for storing quantum programs and sending them for execution over the network via an API.

Note

Qadence currently uses a custom JSON serialization as interchange format. Support for QASM format for digital quantum programs is currently under consideration.

• serialize/deserialize: serialize and deserialize a Qadence object into a dictionary
• save/load: save and load a Qadence object to a file with one of the supported formats. Currently, these are .json and the PyTorch-compatible .pt format.

Let's start with serialization into a dictionary.

import torch
from qadence import QuantumCircuit, QuantumModel, DiffMode
from qadence import chain, hamiltonian_factory, feature_map, hea, Z
from qadence.serialization import serialize, deserialize

n_qubits = 4

my_block = chain(feature_map(n_qubits, param="x"), hea(n_qubits, depth=2))
obs = hamiltonian_factory(n_qubits, detuning=Z)

# Use the block defined above to create a quantum circuit
# serialize/deserialize it
qc = QuantumCircuit(n_qubits, my_block)
qc_dict = serialize(qc)
qc_deserialized = deserialize(qc_dict)
assert qc == qc_deserialized

# Let's wrap it in a QuantumModel
# and serialize it
qm = QuantumModel(qc, obs, diff_mode=DiffMode.AD)
qm_dict = serialize(qm)
qm_deserialized = deserialize(qm_dict)

# Check if the loaded QuantumModel returns the same expectation
values = {"x": torch.rand(10)}
assert torch.allclose(qm.expectation(values=values), qm_deserialized.expectation(values=values))

Finally, we can save the quantum circuit and the model with the two supported formats.

from qadence.serialization import serialize, deserialize, save, load, SerializationFormat

qc_fname = "circuit"
save(qc, folder=".", file_name=qc_fname, format=SerializationFormat.PT)
loaded_qc = load(f"{qc_fname}.pt")
assert qc == loaded_qc

qm_fname = "model"
save(qm, folder=".", file_name=qm_fname, format=SerializationFormat.JSON)
model = load(f"{qm_fname}.json")
assert isinstance(model, QuantumModel)