paddle_quantum.data_analysis.vqls

The VQLS model.

paddle_quantum.data_analysis.vqls.hadamard_test(phi, U, num_qubits)

Given unitary U and state \(|\phi\rangle\), it computes the expectation of U with respect to \(|\phi\rangle\), i.e. \(|\phi\rangle\).

Parameters:

Returns:

Return the real and imaginary part of the expectation value.

Return type:

Tuple[Tensor, Tensor]

paddle_quantum.data_analysis.vqls.hadamard_overlap_test(phi, b, An, Am, num_qubits)

Given unitary Am, An and state \(|\phi\rangle\), b, it computes the value of \(\langle{b}| An |\phi\rangle\langle\phi| Am^\dagger |b\rangle\).

Parameters:

Returns:

Return the real and imaginary part of the calculation.

Return type:

Tuple[Tensor, Tensor]

class paddle_quantum.data_analysis.vqls.VQLS(num_qubits, A, coefficients_real, coefficients_img, b, depth)

Bases: Layer

The class of the variational quantum linear solver (VQLS).

Parameters:

num_qubits (int) – The number of qubits which the quantum circuit contains.
A (List[Tensor]) – List of unitaries in the decomposition of the input matrix.
coefficients_real (List[float]) – Real part of coefficients of corresponding unitaries in the decomposition of the input matrix.
coefficients_img (List[float]) – Imaginary part of coefficients of corresponding unitaries in the decomposition of the input matrix.
b (State) – The state which the input answer is encoded into.
depth (int) – Depth of the ansatz circuit.

forward()

The forward function.

paddle_quantum.data_analysis.vqls.compute(A, b, depth, iterations, LR, gamma=0)

Solve the linear equation Ax=b.

Parameters:

A (ndarray) – Input matrix.
b (ndarray) – Input vector.
depth (int) – Depth of ansatz circuit.
iterations (int) – Number of iterations for optimization.
LR (float) – Learning rate of optimizer.
gamma (float | None) – Extra option to end optimization early if loss is below this value. Default to ‘0’.

Returns:

Return the vector x that solves Ax=b.

Raises:

Return type:

ndarray