def update(
self,
transitions: List[Transition],
model_weights: np.ndarray,
t: int = None,
):
test_X = [trans.X for trans in transitions]
test_w = [trans.w for trans in transitions]
reference_nr_part = self.nr_particles
target_cv = self.mean_cv
cv_estimate = predict_population_size(
reference_nr_part,
target_cv,
lambda nr_particles: calc_cv(
nr_particles,
model_weights,
self.n_bootstrap,
test_w,
transitions,
test_X,
)[0],
)
if not np.isnan(cv_estimate.n_estimated):
self.nr_particles = max(
min(int(cv_estimate.n_estimated), self.max_population_size),
self.min_population_size,
)
logger.info("Change nr particles {} -> {}".format(
reference_nr_part, self.nr_particles))
def adapt_population_size(self, transitions: List[Transition],
model_weights: np.ndarray):
test_X = [trans.X for trans in transitions]
test_w = [trans.w for trans in transitions]
N_BOOTSTR = 5
reference_nr_part = self.nr_particles
target_cv = self.mean_cv
cv_estimate = predict_population_size(
reference_nr_part, target_cv, lambda nr_particles: calc_cv(
nr_particles, model_weights, N_BOOTSTR, test_w, transitions,
test_X)[0])
if not np.isnan(cv_estimate.n_estimated):
self.nr_particles = max(
min(int(cv_estimate.n_estimated), self.max_population_size),
self.min_population_size)
adaptation_logger.info("Change nr particles {} -> {}".format(
reference_nr_part, self.nr_particles))