sklearn.cross_decomposition
.PLSCanonical¶

class
sklearn.cross_decomposition.
PLSCanonical
(n_components=2, *, scale=True, algorithm='nipals', max_iter=500, tol=1e06, copy=True)[source]¶ Partial Least Squares transformer and regressor.
Read more in the User Guide.
New in version 0.8.
 Parameters
 n_componentsint, default=2
Number of components to keep. Should be in
[1, min(n_samples, n_features, n_targets)]
. scalebool, default=True
Whether to scale
X
andY
. algorithm{‘nipals’, ‘svd’}, default=’nipals’
The algorithm used to estimate the first singular vectors of the crosscovariance matrix. ‘nipals’ uses the power method while ‘svd’ will compute the whole SVD.
 max_iterint, default=500
the maximum number of iterations of the power method when
algorithm='nipals'
. Ignored otherwise. tolfloat, default=1e06
The tolerance used as convergence criteria in the power method: the algorithm stops whenever the squared norm of
u_i  u_{i1}
is less thantol
, whereu
corresponds to the left singular vector. copybool, default=True
Whether to copy
X
andY
in fit before applying centering, and potentially scaling. If False, these operations will be done inplace, modifying both arrays.
 Attributes
 x_weights_ndarray of shape (n_features, n_components)
The left singular vectors of the crosscovariance matrices of each iteration.
 y_weights_ndarray of shape (n_targets, n_components)
The right singular vectors of the crosscovariance matrices of each iteration.
 x_loadings_ndarray of shape (n_features, n_components)
The loadings of
X
. y_loadings_ndarray of shape (n_targets, n_components)
The loadings of
Y
. x_scores_ndarray of shape (n_samples, n_components)
The transformed training samples.
Deprecated since version 0.24:
x_scores_
is deprecated in 0.24 and will be removed in 1.1 (renaming of 0.26). You can just calltransform
on the training data instead. y_scores_ndarray of shape (n_samples, n_components)
The transformed training targets.
Deprecated since version 0.24:
y_scores_
is deprecated in 0.24 and will be removed in 1.1 (renaming of 0.26). You can just calltransform
on the training data instead. x_rotations_ndarray of shape (n_features, n_components)
The projection matrix used to transform
X
. y_rotations_ndarray of shape (n_features, n_components)
The projection matrix used to transform
Y
. coef_ndarray of shape (n_features, n_targets)
The coefficients of the linear model such that
Y
is approximated asY = X @ coef_
. n_iter_list of shape (n_components,)
Number of iterations of the power method, for each component. Empty if
algorithm='svd'
.
Examples
>>> from sklearn.cross_decomposition import PLSCanonical >>> X = [[0., 0., 1.], [1.,0.,0.], [2.,2.,2.], [2.,5.,4.]] >>> Y = [[0.1, 0.2], [0.9, 1.1], [6.2, 5.9], [11.9, 12.3]] >>> plsca = PLSCanonical(n_components=2) >>> plsca.fit(X, Y) PLSCanonical() >>> X_c, Y_c = plsca.transform(X, Y)
Methods
fit
(X, Y)Fit model to data.
fit_transform
(X[, y])Learn and apply the dimension reduction on the train data.
get_params
([deep])Get parameters for this estimator.
Transform data back to its original space.
predict
(X[, copy])Predict targets of given samples.
score
(X, y[, sample_weight])Return the coefficient of determination \(R^2\) of the prediction.
set_params
(**params)Set the parameters of this estimator.
transform
(X[, Y, copy])Apply the dimension reduction.

fit
(X, Y)[source]¶ Fit model to data.
 Parameters
 Xarraylike of shape (n_samples, n_features)
Training vectors, where
n_samples
is the number of samples andn_features
is the number of predictors. Yarraylike of shape (n_samples,) or (n_samples, n_targets)
Target vectors, where
n_samples
is the number of samples andn_targets
is the number of response variables.

fit_transform
(X, y=None)[source]¶ Learn and apply the dimension reduction on the train data.
 Parameters
 Xarraylike of shape (n_samples, n_features)
Training vectors, where n_samples is the number of samples and n_features is the number of predictors.
 yarraylike of shape (n_samples, n_targets), default=None
Target vectors, where n_samples is the number of samples and n_targets is the number of response variables.
 Returns
 x_scores if Y is not given, (x_scores, y_scores) otherwise.

get_params
(deep=True)[source]¶ Get parameters for this estimator.
 Parameters
 deepbool, default=True
If True, will return the parameters for this estimator and contained subobjects that are estimators.
 Returns
 paramsdict
Parameter names mapped to their values.

inverse_transform
(X)[source]¶ Transform data back to its original space.
 Parameters
 Xarraylike of shape (n_samples, n_components)
New data, where
n_samples
is the number of samples andn_components
is the number of pls components.
 Returns
 x_reconstructedarraylike of shape (n_samples, n_features)
Notes
This transformation will only be exact if
n_components=n_features
.

predict
(X, copy=True)[source]¶ Predict targets of given samples.
 Parameters
 Xarraylike of shape (n_samples, n_features)
Samples.
 copybool, default=True
Whether to copy
X
andY
, or perform inplace normalization.
Notes
This call requires the estimation of a matrix of shape
(n_features, n_targets)
, which may be an issue in high dimensional space.

score
(X, y, sample_weight=None)[source]¶ Return the coefficient of determination \(R^2\) of the prediction.
The coefficient \(R^2\) is defined as \((1  \frac{u}{v})\), where \(u\) is the residual sum of squares
((y_true  y_pred) ** 2).sum()
and \(v\) is the total sum of squares((y_true  y_true.mean()) ** 2).sum()
. The best possible score is 1.0 and it can be negative (because the model can be arbitrarily worse). A constant model that always predicts the expected value ofy
, disregarding the input features, would get a \(R^2\) score of 0.0. Parameters
 Xarraylike of shape (n_samples, n_features)
Test samples. For some estimators this may be a precomputed kernel matrix or a list of generic objects instead with shape
(n_samples, n_samples_fitted)
, wheren_samples_fitted
is the number of samples used in the fitting for the estimator. yarraylike of shape (n_samples,) or (n_samples, n_outputs)
True values for
X
. sample_weightarraylike of shape (n_samples,), default=None
Sample weights.
 Returns
 scorefloat
\(R^2\) of
self.predict(X)
wrt.y
.
Notes
The \(R^2\) score used when calling
score
on a regressor usesmultioutput='uniform_average'
from version 0.23 to keep consistent with default value ofr2_score
. This influences thescore
method of all the multioutput regressors (except forMultiOutputRegressor
).

set_params
(**params)[source]¶ Set the parameters of this estimator.
The method works on simple estimators as well as on nested objects (such as
Pipeline
). The latter have parameters of the form<component>__<parameter>
so that it’s possible to update each component of a nested object. Parameters
 **paramsdict
Estimator parameters.
 Returns
 selfestimator instance
Estimator instance.

transform
(X, Y=None, copy=True)[source]¶ Apply the dimension reduction.
 Parameters
 Xarraylike of shape (n_samples, n_features)
Samples to transform.
 Yarraylike of shape (n_samples, n_targets), default=None
Target vectors.
 copybool, default=True
Whether to copy
X
andY
, or perform inplace normalization.
 Returns
x_scores
ifY
is not given,(x_scores, y_scores)
otherwise.