btb.tuning.metamodels.gaussian_process module¶
-
class
btb.tuning.metamodels.gaussian_process.
GaussianCopulaProcessMetaModel
[source]¶ Bases:
btb.tuning.metamodels.gaussian_process.GaussianProcessMetaModel
GaussianCopulaProcessMetaModel class.
This class represents a meta-model using an underlying
GaussianProcessRegressor
fromsklearn.gaussian_process
applyingcopulas.univariate.Univariate
transformations to the input data and afterwards reverts it for the predictions.During the
fit
process, this metamodel trains a univariate copula for each hyperparameter to then compute the cumulative distribution of these. Once the cumulative distribution has been calculated, we calculate the inverse of the normal cumulative distribution usingscipy.stats.norm
and use these transformations to train theGaussianProcessRegressor
model.When predicting the output value, an inverse of the normal cumulative distribution is computed to the normal cumulative distribution, using the previously trained univariate copula with the input data of the score.
-
_MODEL_KWARGS
¶ Dictionary with the default
kwargs
for theGaussianProcessRegressor
instantiation.- Type
dict
-
_MODEL_CLASS
¶ Class to be instantiated and used for the
self._model
instantiation. In this casesklearn.gaussian_process.GaussainProcessRegressor
- Type
type
-
-
class
btb.tuning.metamodels.gaussian_process.
GaussianProcessMetaModel
[source]¶ Bases:
btb.tuning.metamodels.base.BaseMetaModel
GaussianProcessMetaModel class.
This class represents a meta-model using an underlying
GaussianProcessRegressor
fromsklearn.gaussian_process
.-
_MODEL_KWARGS
¶ Dictionary with the default
kwargs
for theGaussianProcessRegressor
instantiation.- Type
dict
-
_MODEL_CLASS
¶ Class to be instantiated and used for the
self._model
instantiation. In this casesklearn.gaussian_process.GaussainProcessRegressor
- Type
type
-