mindquantum.algorithm.library.qjpeg

mindquantum.algorithm.library.qjpeg(n_qubits: int, m_qubits: int)[source]

Construct the circuit for compressing quantum figure with the QJPEG algorithm.

Parameters

n_qubits (int) – The number of qubits used to encode the quantum figure to be compressed.
m_qubits (int) – The number of qubits used to encode the compressed quantum figure.

Note

The input arguments, n_qubits and m_qubits, should both be even, and the n_qubits must be not less than the m_qubits. Please refer to arXiv:2306.09323v2 for more information.

Returns

Circuit, The QJPEG circuit for quantum image compression
List[int], List of indices for remainder qubits that carry the compressed quantum image information
List[int], List of indices for discarded qubits

Examples

>>> from mindquantum import Simulator, normalize
>>> import numpy as np
>>> n_qubits = 4
>>> m_qubits = 2
>>> circ, remainder_qubits, discard_qubits = qjpeg(n_qubits, m_qubits)
>>> print(remainder_qubits, discard_qubits)
[0, 2] [1, 3]
>>> data = np.array([[1,0,0,0], [1,1,0,0], [1,1,1,0], [1,1,1,1]])
>>> state = normalize(data.reshape(-1))
>>> sim = Simulator('mqmatrix', n_qubits)
>>> sim.set_qs(state)
>>> sim.apply_circuit(circ)
>>> rho = sim.get_partial_trace(discard_qubits)
>>> sub_probs = rho.diagonal().real
>>> new_data = sub_probs.reshape((2**(m_qubits//2), -1))
>>> print(new_data)
[[0.3, 0.],
 [0.4, 0.3]]