ENH Add routing to LogisticRegressionCV #26525
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -27,17 +27,28 @@ | |
| from ..preprocessing import LabelBinarizer, LabelEncoder | ||
| from ..svm._base import _fit_liblinear | ||
| from ..utils import ( | ||
| Bunch, | ||
| check_array, | ||
| check_consistent_length, | ||
| check_random_state, | ||
| compute_class_weight, | ||
| ) | ||
| from ..utils._param_validation import Interval, StrOptions | ||
| from ..utils.extmath import row_norms, softmax | ||
| from ..utils.metadata_routing import ( | ||
| MetadataRouter, | ||
| MethodMapping, | ||
| _routing_enabled, | ||
| process_routing, | ||
| ) | ||
| from ..utils.multiclass import check_classification_targets | ||
| from ..utils.optimize import _check_optimize_result, _newton_cg | ||
| from ..utils.parallel import Parallel, delayed | ||
| from ..utils.validation import ( | ||
| _check_method_params, | ||
| _check_sample_weight, | ||
| check_is_fitted, | ||
| ) | ||
| from ._base import BaseEstimator, LinearClassifierMixin, SparseCoefMixin | ||
| from ._glm.glm import NewtonCholeskySolver | ||
| from ._linear_loss import LinearModelLoss | ||
|
|
@@ -576,23 +587,25 @@ def _log_reg_scoring_path( | |
| y, | ||
| train, | ||
| test, | ||
| *, | ||
| pos_class, | ||
| Cs, | ||
| scoring, | ||
| fit_intercept, | ||
| max_iter, | ||
| tol, | ||
| class_weight, | ||
| verbose, | ||
| solver, | ||
| penalty, | ||
| dual, | ||
| intercept_scaling, | ||
| multi_class, | ||
| random_state, | ||
| max_squared_sum, | ||
| sample_weight, | ||
| l1_ratio, | ||
| score_params, | ||
| ): | ||
| """Computes scores across logistic_regression_path | ||
|
|
||
|
|
@@ -704,6 +717,9 @@ def _log_reg_scoring_path( | |
| to using ``penalty='l1'``. For ``0 < l1_ratio <1``, the penalty is a | ||
| combination of L1 and L2. | ||
|
|
||
| score_params : dict | ||
| Parameters to pass to the `score` method of the underlying scorer. | ||
|
|
||
| Returns | ||
| ------- | ||
| coefs : ndarray of shape (n_cs, n_features) or (n_cs, n_features + 1) | ||
|
|
@@ -784,7 +800,9 @@ def _log_reg_scoring_path( | |
| if scoring is None: | ||
| scores.append(log_reg.score(X_test, y_test)) | ||
| else: | ||
| score_params = score_params or {} | ||
| score_params = _check_method_params(X=X, params=score_params, indices=test) | ||
| scores.append(scoring(log_reg, X_test, y_test, **score_params)) | ||
|
|
||
| return coefs, Cs, np.array(scores), n_iter | ||
|
|
||
|
|
@@ -1747,7 +1765,7 @@ def __init__( | |
| self.l1_ratios = l1_ratios | ||
|
|
||
| @_fit_context(prefer_skip_nested_validation=True) | ||
| def fit(self, X, y, sample_weight=None, **params): | ||
| """Fit the model according to the given training data. | ||
|
|
||
| Parameters | ||
|
|
@@ -1763,11 +1781,22 @@ def fit(self, X, y, sample_weight=None): | |
| Array of weights that are assigned to individual samples. | ||
| If not provided, then each sample is given unit weight. | ||
|
|
||
| **params : dict | ||
| Parameters to pass to the underlying splitter and scorer. | ||
|
|
||
| .. versionadded:: 1.4 | ||
|
|
||
| Returns | ||
| ------- | ||
| self : object | ||
| Fitted LogisticRegressionCV estimator. | ||
| """ | ||
| if params and not _routing_enabled(): | ||
| raise ValueError( | ||
| "params is only supported if enable_metadata_routing=True." | ||
| " See the User Guide for more information." | ||
|
There was a problem hiding this comment. Would be useful to get the link to the user guide. There was a problem hiding this comment. I mean this page: https://scikit-learn.org/stable/metadata_routing.html There was a problem hiding this comment. I think it will definitely be useful. But if we add it here in one place whereas there are a number of other cases like this (this particular ValueError) that don't contain this link, won't that be a bit inconsistent? There was a problem hiding this comment. We can open a subsequent PR to improve the consistency. |
||
| ) | ||
|
|
||
| solver = _check_solver(self.solver, self.penalty, self.dual) | ||
|
|
||
| if self.penalty == "elasticnet": | ||
|
|
@@ -1829,9 +1858,23 @@ def fit(self, X, y, sample_weight=None): | |
| else: | ||
| max_squared_sum = None | ||
|
|
||
| if _routing_enabled(): | ||
| routed_params = process_routing( | ||
| obj=self, | ||
| method="fit", | ||
| sample_weight=sample_weight, | ||
| other_params=params, | ||
| ) | ||
| else: | ||
| routed_params = Bunch() | ||
| routed_params.splitter = Bunch(split={}) | ||
| routed_params.scorer = Bunch(score=params) | ||
| if sample_weight is not None: | ||
| routed_params.scorer.score["sample_weight"] = sample_weight | ||
|
|
||
| # init cross-validation generator | ||
| cv = check_cv(self.cv, y, classifier=True) | ||
| folds = list(cv.split(X, y, **routed_params.splitter.split)) | ||
|
|
||
| # Use the label encoded classes | ||
| n_classes = len(encoded_labels) | ||
|
|
@@ -1898,6 +1941,7 @@ def fit(self, X, y, sample_weight=None): | |
| max_squared_sum=max_squared_sum, | ||
| sample_weight=sample_weight, | ||
| l1_ratio=l1_ratio, | ||
| score_params=routed_params.scorer.score, | ||
| ) | ||
| for label in iter_encoded_labels | ||
| for train, test in folds | ||
|
|
@@ -2078,7 +2122,7 @@ def fit(self, X, y, sample_weight=None): | |
|
|
||
| return self | ||
|
|
||
| def score(self, X, y, sample_weight=None, **score_params): | ||
| """Score using the `scoring` option on the given test data and labels. | ||
|
|
||
| Parameters | ||
|
|
@@ -2092,15 +2136,74 @@ def score(self, X, y, sample_weight=None): | |
| sample_weight : array-like of shape (n_samples,), default=None | ||
| Sample weights. | ||
|
|
||
| **score_params : dict | ||
| Parameters to pass to the `score` method of the underlying scorer. | ||
|
|
||
| .. versionadded:: 1.4 | ||
|
|
||
| Returns | ||
| ------- | ||
| score : float | ||
| Score of self.predict(X) w.r.t. y. | ||
| """ | ||
| if score_params and not _routing_enabled(): | ||
| raise ValueError( | ||
| "score_params is only supported if enable_metadata_routing=True." | ||
| " See the User Guide for more information." | ||
| " https://scikit-learn.org/stable/metadata_routing.html" | ||
| ) | ||
|
|
||
| scoring = self._get_scorer() | ||
adrinjalali marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| if _routing_enabled(): | ||
| routed_params = process_routing( | ||
| obj=self, | ||
| method="score", | ||
| sample_weight=sample_weight, | ||
| other_params=score_params, | ||
| ) | ||
| else: | ||
| routed_params = Bunch() | ||
| routed_params.scorer = Bunch(score={}) | ||
| if sample_weight is not None: | ||
| routed_params.scorer.score["sample_weight"] = sample_weight | ||
|
|
||
| return scoring( | ||
adrinjalali marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| self, | ||
| X, | ||
| y, | ||
| **routed_params.scorer.score, | ||
| ) | ||
|
|
||
| def get_metadata_routing(self): | ||
| """Get metadata routing of this object. | ||
|
|
||
| Please check :ref:`User Guide <metadata_routing>` on how the routing | ||
| mechanism works. | ||
|
|
||
| .. versionadded:: 1.4 | ||
|
|
||
| Returns | ||
| ------- | ||
| routing : MetadataRouter | ||
| A :class:`~utils.metadata_routing.MetadataRouter` encapsulating | ||
| routing information. | ||
| """ | ||
|
|
||
| router = ( | ||
| MetadataRouter(owner=self.__class__.__name__) | ||
| .add_self_request(self) | ||
| .add( | ||
| splitter=self.cv, | ||
| method_mapping=MethodMapping().add(callee="split", caller="fit"), | ||
| ) | ||
| .add( | ||
| scorer=self._get_scorer(), | ||
| method_mapping=MethodMapping() | ||
| .add(callee="score", caller="score") | ||
| .add(callee="score", caller="fit"), | ||
| ) | ||
| ) | ||
| return router | ||
|
|
||
| def _more_tags(self): | ||
| return { | ||
|
|
@@ -2110,3 +2213,10 @@ def _more_tags(self): | |
| ), | ||
| } | ||
| } | ||
|
|
||
| def _get_scorer(self): | ||
| """Get the scorer based on the scoring method specified. | ||
| The default scoring method is `accuracy`. | ||
| """ | ||
| scoring = self.scoring or "accuracy" | ||
| return get_scorer(scoring) | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
With the removal of the default parameters, the docstring would need their removal too.