def nucmin_estimator(A, y, eta=None, **kwargs):
"""@todo: Docstring for nucmin_estimator.
:param A: @todo
:param y: @todo
:param **kwargs: @todo
:returns: @todo
"""
x_sharp = cvx.Variable(A.shape[1], A.shape[2])
objective = cvx.Minimize(cvx.normNuc(x_sharp))
if eta is None:
constraints = [_expval(A, x_sharp) == y]
else:
constraints = [cvx.abs(_expval(A, x_sharp) - y) < eta]
problem = cvx.Problem(objective, constraints)
problem.solve(**kwargs)
if problem.status not in ['optimal']:
raise ValueError("Optimization did not converge: " + problem.status)
return np.array(x_sharp.value)
def test_norm_nuc(self) -> None:
"""Test gradient for norm_nuc
"""
expr = cp.normNuc(self.A)
self.A.value = [[10, 4], [4, 30]]
self.assertItemsAlmostEqual(expr.grad[self.A].toarray(), [1, 0, 0, 1])
# Set-up hyperparameters
options = {'c1': 0.5, 'c2': 0.3, 'w': 0.9}
# Call instance of PSO
optimizer = ps.single.GlobalBestPSO(n_particles=100,
dimensions=p,
options=options)
# Perform optimization
best_cost, best_pos = optimizer.optimize(nnm,
iters=100,
Y=Y_low,
omega_c=omega_c)
known_value_indices = tuple(zip(*omega.tolist()))
known_values = Y_low[omega[:, 0], omega[:, 1]]
X = cp.Variable((m, n), pos=True)
objective_fn = cp.normNuc(X)
constraints = [
X[known_value_indices] == known_values,
]
problem = cp.Problem(cp.Minimize(objective_fn), constraints)
problem.solve(gp=False)
#print("Solver reconstruct loss: ", np.linalg.norm((np.array(X.value) - Y_low), "fro"))
Y[omega_c[:, 0], omega_c[:, 1]] = best_pos
print("PSO reconstruct loss:", np.linalg.norm((Y - Y_low), "fro"))
result.append(
[t,
np.linalg.norm((Y - Y_low), "fro") / np.linalg.norm(Y_low, "fro")])
#result.append([t, "CVXPY", np.linalg.norm((np.array(X.value) - Y_low), "fro")])
df = pd.DataFrame(result, columns=["iter", "MSE"]).iloc[:, 1]