clearbox/models/probabilistic.py [295:335]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
  def __init__(
      self,
      surrogate_model: SurrogateModel,
      acquisition_function: AcquisitionFunction,
      metric: cbx_metrics.BaseMetric,
      n_opt_steps: int,
      batch_size: int,
      seed_batch_size: int,
      grid_size: int,
      print_progress: bool = False,
  ):
    """Bayesian Optimization over linear model constructor.

    Args:
      surrogate_model: The surrogate model builder object to use for mean and
        standard deviation predictions.
      acquisition_function: Acquisition function object which is used to pick
        the next candidate points to compute the optimized function for.
      metric: The metric to optimize, the values of the metric are used as
        targets in the optimization process.
      n_opt_steps: Number of the optimization steps to perform.
      batch_size: Number of candidates pick on each optimization step.
      seed_batch_size: Number of the candidates to compute the `metric` value
        for before the first optimization step. Those values are then used as
        the initial training set for the `surrogate_model`.
      grid_size: The number of steps to use for single feature in the grid. In
        the canonical implementation of BO the candidates are sampled from the
        multivariate normal distribution at each optimization step. But, as our
        implementation is basically a logical evolution of Grid Search
        algorithm, we retain the same uniform grid and use it as a source of
        candidates.
      print_progress: Whether to print the progress of the optimization.
    """
    self._surrogate_model = surrogate_model
    self._acquisition_function = acquisition_function
    self._metric = metric
    self._n_opt_steps = n_opt_steps
    self._batch_size = batch_size
    self._seed_batch_size = seed_batch_size
    self._grid_size = grid_size
    self._print_progress = print_progress
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



clearbox/models/probabilistic.py [454:499]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
  def __init__(
      self,
      surrogate_model: SurrogateModel,
      acquisition_function: AcquisitionFunction,
      metric: cbx_metrics.BaseMetric,
      n_opt_steps: int,
      batch_size: int,
      seed_batch_size: int,
      grid_size: int,
      print_progress: bool = False,
  ):
    """Constructor of the Bayesian Optimization model builder.

    Args:
      surrogate_model:
        The surrogate model builder object to use for mean and standard
        deviation predictions.
      acquisition_function:
        Acquisition function object which is used to pick the next candidate
        points to compute the optimized function for.
      metric:
        The metric to optimize, the values of the metric are used as targets in
        the optimization process.
      n_opt_steps: Number of the optimization steps to perform.
      batch_size: Number of candidates pick on each optimization step.
      seed_batch_size:
        Number of the candidates to compute the `metric` value for before the
        first optimization step. Those values are then used as the initial
        training set for the `surrogate_model`.
      grid_size:
        The number of steps to use for single feature in the grid. In the
        canonical implementation of BO the candidates are sampled from the
        multivariate normal distribution at each optimization step. But, as our
        implementation is basically a logical evolution of Grid Search
        algorithm, we retrain the same uniform grid and use it as a source of
        candidates.
      print_progress: Whether to print the progress of the optimization.
    """
    self._surrogate_model = surrogate_model
    self._acquisition_function = acquisition_function
    self._metric = metric
    self._n_opt_steps = n_opt_steps
    self._batch_size = batch_size
    self._seed_batch_size = seed_batch_size
    self._grid_size = grid_size
    self._print_progress = print_progress
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -