gpCAM

class gen_classes.gpCAM.GP_CAM(vocs, ask_max_iter=10, random_seed=1, *args, **kwargs)

Bases: LibensembleGenerator

This generation function constructs a global surrogate of f values.

It is a batched method that produces a first batch uniformly random from (lb, ub). On subsequent iterations, it calls an optimization method to produce the next batch of points. This optimization might be too slow (relative to the simulation evaluation time) for some use cases.

Parameters:

  • vocs (VOCS)

  • ask_max_iter (int)

  • random_seed (int)

suggest(num_points=0)

Request the next set of points to evaluate.

Parameters:

num_points (int | None)

Return type:

List[dict]

ingest(results)

Send the results of evaluations to the generator.

Parameters:

results (List[dict])

Return type:

None

class gen_classes.gpCAM.GP_CAM_Covar(vocs, test_points_file=None, use_grid=False, *args, **kwargs)

Bases: GP_CAM

This generation function constructs a global surrogate of f values.

It is a batched method that produces a first batch uniformly random from (lb, ub) and on following iterations samples the GP posterior covariance function to find sample points.

Parameters:

  • vocs (VOCS)

  • test_points_file (str)

  • use_grid (bool)

suggest(num_points=0)

Request the next set of points to evaluate.

Parameters:

num_points (int | None)

Return type:

List[dict]

ingest(results)

Send the results of evaluations to the generator.

Parameters:

results (List[dict])

Return type:

None

See also

 1    vocs = VOCS(variables={"x0": [-3, 3], "x1": [-2, 2], "x2": [-1, 1], "x3": [-1, 1]}, objectives={"f": "MINIMIZE"})
 2
 3    gen = GP_CAM_Covar(vocs)
 4
 5    for inst in range(3):
 6        if inst == 0:
 7            gen_specs["generator"] = gen
 8            num_batches = 10
 9            exit_criteria = {"sim_max": num_batches * batch_size, "wallclock_max": 300}
10            libE_specs["save_every_k_gens"] = 150
11            libE_specs["H_file_prefix"] = "gpCAM_nongrid"
12        if inst == 1:
13            gen = GP_CAM_Covar(vocs, use_grid=True, test_points_file="gpCAM_nongrid_after_gen_150.npy")
14            gen_specs["generator"] = gen
15            libE_specs["final_gen_send"] = True
16            del libE_specs["H_file_prefix"]
17            del libE_specs["save_every_k_gens"]
18        elif inst == 2:
19            gen = GP_CAM(vocs, ask_max_iter=1)
20            gen_specs["generator"] = gen
21            num_batches = 3  # Few because the ask_tell gen can be slow
22            exit_criteria = {"sim_max": num_batches * batch_size, "wallclock_max": 300}
23
24        # Perform the run
25        H, persis_info, flag = libE(sim_specs, gen_specs, exit_criteria, {}, libE_specs=libE_specs)