def main():
""" Main Function. """
# Choose which objective to minimise and the configuration file
objective_to_min, config_file, fidel_cost_func = _CHOOSER_DICT[(PROBLEM,
IS_MF)]
config = load_config_file(config_file)
log_stream = open(LOG_FILE, 'w')
# Call the optimiser
if IS_MF:
opt_val, opt_pt, history = minimise_multifidelity_function(
objective_to_min,
config.fidel_space,
config.domain,
config.fidel_to_opt,
fidel_cost_func,
MAX_CAPITAL,
capital_type='realtime',
config=config,
reporter=log_stream)
else:
opt_val, opt_pt, history = minimise_function(objective_to_min,
config.domain,
MAX_CAPITAL,
capital_type='realtime',
config=config,
reporter=log_stream)
# Print out result
log_stream.close()
print('Optimum Value found in %02.f time (%d evals): %0.4f' %
(MAX_CAPITAL, len(history.curr_opt_points), opt_val))
print('Optimum Point: %s.' % (str(opt_pt)))
def run_branin_benchmarks(rep):
experiment, f = benchmark_minimize_callable(
problem=branin_100,
num_trials=50,
method_name='add_gp_ucb',
replication_index=rep,
)
options = Namespace(acq="add_ucb")
res = minimise_function(
f,
domain=[[-5, 10]] * 50 + [[0, 15]] * 50,
max_capital=49,
options=options,
)
with open(f'results/branin_100_addgpucb_rep_{rep}.json', "w") as fout:
json.dump(object_to_json(experiment), fout)
def minimize_function(self, f, n_iter=10, verbose=True, seed=None):
"""
Run Dragonfly Bayesian optimization to minimize function f.
Parameters
----------
f : function
Function to optimize.
n_iter : int
Number of iterations of Bayesian optimization.
verbose : bool
If True, print information.
seed : int
If not None, set the random seed to seed.
"""
if seed is not None:
np.random.seed(seed)
domain = self._get_domain()
opt_method = 'bo'
parsed_config = self._get_parsed_config()
options = self._get_options()
opt_val, opt_pt, history = dragonfly.minimise_function(
func=f,
domain=domain,
max_capital=n_iter,
opt_method=opt_method,
config=parsed_config,
options=options,
)
results = Namespace(opt_val=opt_val, opt_pt=opt_pt, history=history)
if verbose:
vals = results.history.query_vals
min_vals_idx = min(range(len(vals)), key=lambda x: vals[x])
print('Minimum y = {}'.format(results.opt_val))
print('Minimizer x = {}'.format(results.opt_pt))
print('Found at iter = {}'.format(min_vals_idx + 1))
return results
def run_hartmann6_benchmarks(D, rep):
if D == 100:
problem = hartmann6_100
elif D == 1000:
problem = hartmann6_1000
experiment, f = benchmark_minimize_callable(
problem=problem,
num_trials=200,
method_name='add_gp_ucb',
replication_index=rep,
)
options = Namespace(acq="add_ucb")
res = minimise_function(f,
domain=[[0, 1]] * D,
max_capital=199,
options=options)
with open(f'results/hartmann6_{D}_addgpucb_rep_{rep}.json', "w") as fout:
json.dump(object_to_json(experiment), fout)