nltools.cross_validation¶

The cross-validation splitter nltools adds to the scikit-learn ones.

Scikit-learn-compatible cross-validation data classes.

Classes:

Name	Description
`KFoldStratified`	Stratify continuous targets across K-fold cross-validation.

Classes¶

KFoldStratified ¶

KFoldStratified(
    n_splits=3, *, shuffle=False, random_state=None
)

Bases: _BaseKFold

Stratify continuous targets across K-fold cross-validation.

Unlike the scikit-learn equivalent, this iterator stratifies continuous data.

Provides train/test indices to split data in train test sets. Samples are ordered by their continuous target y and dealt round-robin into k folds so each fold spans the full range of y. Each fold is then used as a validation set once while the k - 1 remaining folds form the training set.

Parameters:

Name	Type	Description	Default
`n_splits`	`int`	Number of folds. Must be at least 2. Defaults to 3.	`3`
`shuffle`	`bool`	Whether to break ties in `y` randomly before dealing samples into folds. Default False.	`False`
`random_state`	`int \| RandomState`	Seed or RandomState for the tie-break shuffle. If None, use the default numpy RNG.	`None`

Methods:

Name	Description
`split`	Generate indices to split data into training and test set.

Methods:¶

split ¶

split(X, y=None, groups=None)

Generate indices to split data into training and test set.

Parameters:

Name	Type	Description	Default
`X`	`array - like`	Training data of shape `(n_samples, n_features)`. Only `y` is needed to generate the splits, so `np.zeros(n_samples)` works as a placeholder.	required
`y`	`array - like`	Continuous target of shape `(n_samples,)` or `(n_samples, 1)`; stratification is based on its ordering.	`None`
`groups`	`array - like`	Always ignored; exists for sklearn compatibility.	`None`

Yields:

Type	Description
`tuple[ndarray, ndarray]`	`(train, test)` — the training set indices and the testing set indices for that split.

Raises:

Type	Description
`ValueError`	If `y` carries more than one target per sample. Ordering samples by a multi-target response has no single meaning, and flattening one would produce more fold labels than there are rows.