[QST] how to use python cuopt

[pengfzhou]

for big problem, i want use the local search or heuristic method ,how to use ?
like this:
settings = SolverSettings()
settings.set_parameter("time_limit", time_limit)
settings.set_parameter("mip_heuristics_only", 1)
problem.solve(settings)

[pengfzhou]

I use cuopt to solve a opb problem, it cannot get a solution , I dont know why, below it the output
(base) zpf@huawei:~$ python solve_opb.py ~/pbs/normalized-SameQueensKnights-b-35_c24.opb
Setting parameter time_limit to 3.600000e+03
Setting parameter mip_heuristics_only to true
cuOpt version: 26.6.0, git hash: df5fe0d, host arch: x86_64, device archs: 75-real,80-real,86-real,90a-real,100f-real,120a-real,120
CPU: 12th Gen Intel(R) Core(TM) i9-12900KF, threads (physical/logical): 16/24, RAM: 36.50 GiB
CUDA 13.0, device: NVIDIA GeForce RTX 3080 Ti (ID 0), VRAM: 11.63 GiB
CUDA device UUID: d6601758-c1dc-1657-77a8-e91d4661781c

Solving a problem with 55130 constraints, 43918 variables (43918 integers), and 506232 nonzeros
Problem scaling:
Objective coefficents range: [1e+00, 3e+01]
Constraint matrix coefficients range: [1e+00, 5e+02]
Constraint rhs / bounds range: [0e+00, 5e+02]
Variable bounds range: [0e+00, 1e+00]

MIP row scaling completed
Original problem: 55130 constraints, 43918 variables, 506232 nonzeros
Calling Papilo presolver (git hash 741a2b9c)
Presolve status: reduced the problem
Presolve removed: 31851 constraints, 18382 variables, 105170 nonzeros
Presolved problem: 23279 constraints, 25536 variables (25536 integer), 401062 nonzeros
Papilo presolve time: 3.81
Objective offset 0.000000 scaling_factor 1.000000
Model fingerprint: 0x43c48413
Starting cuOpt presolve
Running probing cache with 23 tasks
New solution from early primal heuristics (CPUFJ). Objective +0.000000e+00. Time 0.23
After cuOpt presolve: 23279 constraints, 25536 variables, objective offset 0.000000.
cuOpt presolve time: 48.77
Graph construction time 0.015246
Automorphism group size 8*10^0, 2 dejavu generators (2 move variables)
Dejavu time 0.054161
Projected 2 generators onto 25536 binary variables (0 skipped non-binary), 2 stored
Binary orbits: 3547 non-trivial, max size 8, 25508/25536 (99.9%) binary variables in orbits
Orbit histogram (size: count):
4:717
8:2830

Total symmetry detection time 0.069510
Objective function is integral, scale 1
Objective step size 1 (unscaled: 1)