Error when the objective function includes a linear term

[eegzheng]

Hello,

I am trying a toy example by adapting the code from your max-cut example and is shown below, but I found as long as there is a linear term in the objective function, it will shows error below:
... \SimCIM.py", line 86, in update
(self.J.dot(self.x) + self.h) * self.dt
ValueError: operands could not be broadcast together with shapes (8,100) (8,)

Without h or without s.update(), it works fine.

##########################################################
import numpy as np
from mindquantum.algorithm.qaia import ASB, BSB, DSB, SimCIM, NMFA

N = 8 # number of spins

QUBO problem formulation

min x'Jx + h'*x, x is a spin vector variable

J = np.array([[ 0, 50, 0, 0, -50, -100, 0, 0],
[ 50, 0, 0, 0, -50, -100, 0, 0],
[ 0, 0, 0, 50, 0, 0, -50, -100],
[ 0, 0, 50, 0, 0, 0, -50, -100],
[ -50, -50, 0, 0, 0, 100, -2, 0],
[-100, -100, 0, 0, 100, 0, 0, 0],
[ 0, 0, -50, -50, -2, 0, 0, 100],
[ 0, 0, -100, -100, 0, 0, 100, 0]])
h = np.array([ 103, 100, 100, 100, -100, -200, -100, -200])

sample_size=100
n_iter_list = [10, 25, 50]
cut_value_list = []
for n_iter in n_iter_list:
s = SimCIM(-2*J,-h,batch_size=sample_size, n_iter=n_iter)
s.update()
cut_value = s.calc_energy()
cut_value_list.append(cut_value)

I also wonder how to extra the desired solution and the associated objective value. Is x = np.sign(s.x) the way to find all sample solutions, and the associated solutions are stored in s.calc_energy()?

Many thanks!