ddl.linear
.BestLinearReconstructionDestructor¶
-
class
ddl.linear.
BestLinearReconstructionDestructor
(linear_estimator=None, destructor=None, linear_projector_kwargs=None)[source]¶ Bases:
ddl.base.CompositeDestructor
Best linear reconstruction after a linear destructor.
Class that converts a linear -> destructor combination into a combination that returns the data to as close to the original space as possible. Essentially, linear -> destructor -> independent Gaussian inverse cdf -> inverse linear -> independent Gaussian cdf. For example, if the linear projector was PCA and the destructor was a independent Gaussian, then this would correspond to ZCA whitening.
Parameters: - linear_estimator : estimator, default=IdentityLinearEstimator
A linear estimator that has either a coef_ attribute or a components_. For example,
sklearn.decomposition.PCA
.- destructor : estimator
Density destructor to use in between linear projections.
- linear_projector_kwargs : dict
Keyword arguments to pass when constructing
ddl.linear.LinearProjector
.
Attributes: - fitted_destructors_ : list
List of fitted (sub)destructors. (Note that these objects are cloned via
sklearn.base.clone
from thedestructors
parameter so as to avoid mutating thedestructors
parameter.)- density_ : estimator
Implicit density of composite destructor.
Methods
create_fitted
(fitted_destructors, \*\*kwargs)Create fitted destructor. fit
(self, X[, y])Fit estimator to X. fit_transform
(self, X[, y])[Placeholder]. get_domain
(self)Get the domain of this destructor. get_params
(self[, deep])Get parameters for this estimator. inverse_transform
(self, X[, y, partial_idx])Apply inverse destructive transformation to X. sample
(self[, n_samples, y, random_state])Sample from composite destructor. score
(self, X[, y, partial_idx])Override super class to allow for partial_idx. score_layers
(self, X[, y, partial_idx])Override super class to allow for partial_idx. score_samples
(self, X[, y, partial_idx])Compute log-likelihood (or log(det(Jacobian))) for each sample. score_samples_layers
(self, X[, y, partial_idx])[Placeholder]. set_params
(self, \*\*params)Set the parameters of this estimator. transform
(self, X[, y, partial_idx])Apply destructive transformation to X. -
__init__
(self, linear_estimator=None, destructor=None, linear_projector_kwargs=None)[source]¶ Initialize self. See help(type(self)) for accurate signature.
-
classmethod
create_fitted
(fitted_destructors, **kwargs)¶ Create fitted destructor.
Parameters: - fitted_destructors : array-like of Destructor
Fitted destructors.
- **kwargs
Other parameters to pass to constructor.
Returns: - fitted_transformer : Transformer
Fitted transformer.
-
fit
(self, X, y=None, **fit_params)¶ Fit estimator to X.
Parameters: - X : array-like, shape (n_samples, n_features)
Training data, where n_samples is the number of samples and n_features is the number of features.
- y : None, default=None
Not used in the fitting process but kept for compatibility.
- fit_params : dict, optional
Optional extra fit parameters.
Returns: - self : estimator
Returns the instance itself.
-
fit_transform
(self, X, y=None, **kwargs)[source]¶ [Placeholder].
Parameters: - X :
- y :
- kwargs :
Returns: - obj : object
-
get_domain
(self)¶ Get the domain of this destructor.
Returns: - domain : array-like, shape (2,) or shape (n_features, 2)
If shape is (2, ), then
domain[0]
is the minimum anddomain[1]
is the maximum for all features. If shape is (n_features, 2), then each feature’s domain (which could be different for each feature) is given similar to the first case.
-
get_params
(self, deep=True)¶ Get parameters for this estimator.
Parameters: - deep : boolean, optional
If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns: - params : mapping of string to any
Parameter names mapped to their values.
-
inverse_transform
(self, X, y=None, partial_idx=None)¶ Apply inverse destructive transformation to X.
Parameters: - X : array-like, shape (n_samples, n_features)
New data, where n_samples is the number of samples and n_features is the number of features.
- y : None, default=None
Not used in the transformation but kept for compatibility.
- partial_idx : list or None, default=None
List of indices of the fitted destructor to use in the transformation. The default of None uses all the fitted destructors. Mainly used for visualization or debugging.
Returns: - X_new : array-like, shape (n_samples, n_features)
Transformed data (possibly only partial transformation).
-
sample
(self, n_samples=1, y=None, random_state=None)¶ Sample from composite destructor.
Nearly the same as
DestructorMixin.sample
but the number of features is found from first fitted destructor to avoid recursion.
-
score
(self, X, y=None, partial_idx=None)¶ Override super class to allow for partial_idx.
-
score_layers
(self, X, y=None, partial_idx=None)¶ Override super class to allow for partial_idx.
-
score_samples
(self, X, y=None, partial_idx=None)¶ Compute log-likelihood (or log(det(Jacobian))) for each sample.
Parameters: - X : array-like, shape (n_samples, n_features)
New data, where n_samples is the number of samples and n_features is the number of features.
- y : None, default=None
Not used but kept for compatibility.
- partial_idx : list or None, default=None
List of indices of the fitted destructor to use in the computing the log likelihood. The default of None uses all the fitted destructors. Mainly used for visualization or debugging.
Returns: - log_likelihood : array, shape (n_samples,)
Log likelihood of each data point in X.
-
score_samples_layers
(self, X, y=None, partial_idx=None)¶ [Placeholder].
Parameters: - X :
- y :
- partial_idx :
Returns: - obj : object
-
set_params
(self, **params)¶ Set the parameters of this estimator.
The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form
<component>__<parameter>
so that it’s possible to update each component of a nested object.Returns: - self
-
transform
(self, X, y=None, partial_idx=None)¶ Apply destructive transformation to X.
Parameters: - X : array-like, shape (n_samples, n_features)
New data, where n_samples is the number of samples and n_features is the number of features.
- y : None, default=None
Not used in the transformation but kept for compatibility.
- partial_idx : list or None, default=None
List of indices of the fitted destructor to use in the transformation. The default of None uses all the fitted destructors. Mainly used for visualization or debugging.
Returns: - X_new : array-like, shape (n_samples, n_features)
Transformed data (possibly only partial transformation).