SurrogateModel

class SurrogateModel

Bases: Generic[_T]

A generic surrogate model class for approximating expensive objective functions.

Type Parameters:

_T: A type variable that must be a subclass of Trainer. This allows the surrogate model to use any specific trainer implementation for training and evaluation.

Methods

__init__

Initializes the surrogate model class.

to_poly

Converts the surrogate model to an Amplify Poly object.

train

Trains the surrogate model on the given dataset.

Attributes

trainer

Returns the trainer used in the model.

variables

Returns the variables used in the model.

classmethod __class_getitem__(params)
__init__(surrogate_variables: PolyArray, trainer: _T) None

Initializes the surrogate model class.

Parameters:
  • surrogate_variables (amplify.PolyArray) – The Amplify SDK variables for the model.

  • trainer (_T) – The trainer to use for the model.

classmethod __init_subclass__(*args, **kwargs)

This method is called when a class is subclassed.

The default implementation does nothing. It may be overridden to extend subclasses.

to_poly() Poly

Converts the surrogate model to an Amplify Poly object.

Returns:

The Amplify Poly representation of the surrogate model.

Return type:

amplify.Poly

train(training_data: Dataset) None

Trains the surrogate model on the given dataset.

Parameters:

training_data (Dataset) – The dataset to train on.

__dict__ = mappingproxy({'__module__': 'amplify_bbopt.surrogate', '__doc__': 'A generic surrogate model class for approximating expensive objective functions.\n\n    Type Parameters:\n        _T: A type variable that must be a subclass of `Trainer`. This allows the surrogate model to use any specific trainer implementation for training and evaluation.\n    ', '__init__': <function SurrogateModel.__init__>, 'variables': <property object>, 'trainer': <property object>, 'train': <function SurrogateModel.train>, 'to_poly': <function SurrogateModel.to_poly>, '__orig_bases__': (typing.Generic[~_T],), '__dict__': <attribute '__dict__' of 'SurrogateModel' objects>, '__weakref__': <attribute '__weakref__' of 'SurrogateModel' objects>, '__parameters__': (~_T,), '__annotations__': {'_surrogate_variables': 'amplify.PolyArray', '_trainer': '_T'}})
__orig_bases__ = (typing.Generic[~_T],)
__parameters__ = (~_T,)
__slots__ = ()
__weakref__

list of weak references to the object (if defined)

property trainer: _T

Returns the trainer used in the model.

property variables: PolyArray

Returns the variables used in the model.