cca_zoo.model_selection

Cross-validated hyperparameter search for multiview models.

---

GridSearchCV

GridSearchCV(
    estimator: BaseEstimator,
    param_grid: dict[str, list[Any]]
    | list[dict[str, list[Any]]],
    cv: int | Any = 5,
    scoring: str | None = None,
    n_jobs: int | None = None,
    refit: bool = True,
    verbose: int = 0,
)

Grid search with cross-validation for multiview CCA models.

Wraps :class:sklearn.model_selection.GridSearchCV to support the list[ArrayLike] interface of cca_zoo models. Views are horizontally stacked before being passed to sklearn and split back inside the wrapped estimator.

Parameters:

Name	Type	Description	Default
estimator	BaseEstimator	A multiview CCA estimator (e.g. :class:~cca_zoo.linear.CCA).	required
param_grid	dict[str, list[Any]] | list[dict[str, list[Any]]]	Dictionary or list of dictionaries with parameter names as keys and lists of parameter settings as values.	required
cv	int | Any	Number of cross-validation folds or a cross-validation splitter. Default is 5.	5
scoring	str | None	Scoring strategy. When None the estimator's default :meth:score method is used.	None
n_jobs	int | None	Number of jobs to run in parallel. Default is None (sequential).	None
refit	bool	Whether to refit the best estimator on the full dataset. Default is True.	True
verbose	int	Verbosity level. Default is 0.	0

Example

import numpy as np from cca_zoo.linear import CCA from cca_zoo.model_selection import GridSearchCV rng = np.random.default_rng(0) X1 = rng.standard_normal((50, 5)) X2 = rng.standard_normal((50, 4)) gs = GridSearchCV( ... CCA(), param_grid={"latent_dimensions": [1, 2]}, cv=2 ... ) gs = gs.fit([X1, X2])

Source code in cca_zoo/model_selection/_search.py

def __init__(
    self,
    estimator: BaseEstimator,
    param_grid: dict[str, list[Any]] | list[dict[str, list[Any]]],
    cv: int | Any = 5,
    scoring: str | None = None,
    n_jobs: int | None = None,
    refit: bool = True,
    verbose: int = 0,
) -> None:
    self.estimator = estimator
    self.param_grid = param_grid
    self.cv = cv
    self.scoring = scoring
    self.n_jobs = n_jobs
    self.refit = refit
    self.verbose = verbose

fit

fit(
    views: list[ArrayLike],
    y: None = None,
    **fit_params: Any,
) -> GridSearchCV

Run grid search with cross-validation on multiview data.

Parameters:

Name	Type	Description	Default
views	list[ArrayLike]	List of arrays, each of shape (n_samples, n_features_i). All arrays must have the same number of rows.	required
y	None	Ignored.	None
**fit_params	Any	Additional keyword arguments forwarded to the estimator's fit method during each fold.	{}

Returns:

Name	Type	Description
self	GridSearchCV	Fitted grid search object.

Source code in cca_zoo/model_selection/_search.py

def fit(
    self,
    views: list[ArrayLike],
    y: None = None,
    **fit_params: Any,
) -> GridSearchCV:
    """Run grid search with cross-validation on multiview data.

    Args:
        views: List of arrays, each of shape (n_samples, n_features_i).
            All arrays must have the same number of rows.
        y: Ignored.
        **fit_params: Additional keyword arguments forwarded to the
            estimator's ``fit`` method during each fold.

    Returns:
        self: Fitted grid search object.
    """
    arrays = [np.asarray(v) for v in views]
    split_indices = [a.shape[1] for a in arrays]
    x_concat = np.hstack(arrays)

    wrapped_estimator = _MultiviewWrapper(
        estimator=self.estimator,
        split_indices=split_indices,
    )
    wrapped_grid = self._make_wrapped_param_grid(split_indices)

    self._inner_cv: skms.GridSearchCV = skms.GridSearchCV(
        estimator=wrapped_estimator,
        param_grid=wrapped_grid,
        cv=self.cv,
        scoring=self.scoring,
        n_jobs=self.n_jobs,
        refit=self.refit,
        verbose=self.verbose,
    )
    self._inner_cv.fit(x_concat, y, **fit_params)

    # Expose the most important attributes at the top level
    self.cv_results_: dict[str, Any] = self._inner_cv.cv_results_
    self.best_score_: float = self._inner_cv.best_score_
    # Strip the estimator__ prefix from best_params_
    raw_best: dict[str, Any] = self._inner_cv.best_params_
    self.best_params_: dict[str, Any] = {
        k[len("estimator__") :]: v for k, v in raw_best.items()
    }
    if self.refit:
        self.best_estimator_: BaseEstimator = (
            self._inner_cv.best_estimator_.estimator_
        )
    return self

transform

transform(views: list[ArrayLike]) -> list[np.ndarray]

Transform multiview data using the best estimator.

Mirrors :meth:sklearn.model_selection.GridSearchCV.transform, forwarding to best_estimator_.

Parameters:

Name	Type	Description	Default
views	list[ArrayLike]	List of arrays, each of shape (n_samples, n_features_i).	required

Returns:

Type	Description
list[ndarray]	List of arrays, each of shape (n_samples, latent_dimensions).

Raises:

Type	Description
AttributeError	If refit=False, so no best_estimator_ was fitted.

Source code in cca_zoo/model_selection/_search.py

def transform(self, views: list[ArrayLike]) -> list[np.ndarray]:
    """Transform multiview data using the best estimator.

    Mirrors :meth:`sklearn.model_selection.GridSearchCV.transform`,
    forwarding to ``best_estimator_``.

    Args:
        views: List of arrays, each of shape (n_samples, n_features_i).

    Returns:
        List of arrays, each of shape (n_samples, latent_dimensions).

    Raises:
        AttributeError: If ``refit=False``, so no ``best_estimator_``
            was fitted.
    """
    if not self.refit:
        raise AttributeError(
            "`transform` is not available when `refit=False`; no "
            "`best_estimator_` was fitted. Set `refit=True` to use it."
        )
    return cast(list[np.ndarray], self.best_estimator_.transform(views))

score

score(views: list[ArrayLike], y: None = None) -> float

Score the best estimator on held-out multiview data.

Parameters:

Name	Type	Description	Default
views	list[ArrayLike]	List of arrays, each of shape (n_samples, n_features_i).	required
y	None	Ignored.	None

Returns:

Name	Type	Description
Scalar	float	mean canonical correlation of the best estimator.

Source code in cca_zoo/model_selection/_search.py

def score(self, views: list[ArrayLike], y: None = None) -> float:
    """Score the best estimator on held-out multiview data.

    Args:
        views: List of arrays, each of shape (n_samples, n_features_i).
        y: Ignored.

    Returns:
        Scalar: mean canonical correlation of the best estimator.
    """
    arrays = [np.asarray(v) for v in views]
    x_concat = np.hstack(arrays)
    return float(self._inner_cv.score(x_concat, y))