MAINT Add parameter validation for LabelPropagation and LabelSpreading (scikit-learn#24211)

Author: stefmolin

sklearn/semi_supervised/_label_propagation.py

@@ -56,6 +56,7 @@
 #          Utkarsh Upadhyay <[email protected]>
 # License: BSD
 from abc import ABCMeta, abstractmethod
+from numbers import Integral, Real
 
 import warnings
 import numpy as np
@@ -68,6 +69,7 @@
 from ..utils.extmath import safe_sparse_dot
 from ..utils.multiclass import check_classification_targets
 from ..utils.validation import check_is_fitted
+from ..utils._param_validation import Interval, StrOptions
 from ..exceptions import ConvergenceWarning
 
 
@@ -105,6 +107,16 @@ class BaseLabelPropagation(ClassifierMixin, BaseEstimator, metaclass=ABCMeta):
          for more details.
     """
 
+    _parameter_constraints: dict = {
+        "kernel": [StrOptions({"knn", "rbf"}), callable],
+        "gamma": [Interval(Real, 0, None, closed="left")],
+        "n_neighbors": [Interval(Integral, 0, None, closed="neither")],
+        "alpha": [None, Interval(Real, 0, 1, closed="neither")],
+        "max_iter": [Interval(Integral, 0, None, closed="neither")],
+        "tol": [Interval(Real, 0, None, closed="left")],
+        "n_jobs": [None, Integral],
+    }
+
     def __init__(
         self,
         kernel="rbf",
@@ -152,13 +164,6 @@ def _get_kernel(self, X, y=None):
                 return self.kernel(X, X)
             else:
                 return self.kernel(X, y)
-        else:
-            raise ValueError(
-                "%s is not a valid kernel. Only rbf and knn"
-                " or an explicit function "
-                " are supported at this time."
-                % self.kernel
-            )
 
     @abstractmethod
     def _build_graph(self):
@@ -246,6 +251,7 @@ def fit(self, X, y):
         self : object
             Returns the instance itself.
         """
+        self._validate_params()
         X, y = self._validate_data(X, y)
         self.X_ = X
         check_classification_targets(y)
@@ -261,14 +267,6 @@ def fit(self, X, y):
 
         n_samples, n_classes = len(y), len(classes)
 
-        alpha = self.alpha
-        if self._variant == "spreading" and (
-            alpha is None or alpha <= 0.0 or alpha >= 1.0
-        ):
-            raise ValueError(
-                "alpha=%s is invalid: it must be inside the open interval (0, 1)"
-                % alpha
-            )
         y = np.asarray(y)
         unlabeled = y == -1
 
@@ -283,7 +281,7 @@ def fit(self, X, y):
             y_static[unlabeled] = 0
         else:
             # LabelSpreading
-            y_static *= 1 - alpha
+            y_static *= 1 - self.alpha
 
         l_previous = np.zeros((self.X_.shape[0], n_classes))
 
@@ -310,7 +308,7 @@ def fit(self, X, y):
             else:
                 # clamp
                 self.label_distributions_ = (
-                    np.multiply(alpha, self.label_distributions_) + y_static
+                    np.multiply(self.alpha, self.label_distributions_) + y_static
                 )
         else:
             warnings.warn(
@@ -417,6 +415,9 @@ class LabelPropagation(BaseLabelPropagation):
 
     _variant = "propagation"
 
+    _parameter_constraints: dict = {**BaseLabelPropagation._parameter_constraints}
+    _parameter_constraints.pop("alpha")
+
     def __init__(
         self,
         kernel="rbf",
@@ -573,6 +574,9 @@ class LabelSpreading(BaseLabelPropagation):
 
     _variant = "spreading"
 
+    _parameter_constraints: dict = {**BaseLabelPropagation._parameter_constraints}
+    _parameter_constraints["alpha"] = [Interval(Real, 0, 1, closed="neither")]
+
     def __init__(
         self,
         kernel="rbf",

sklearn/semi_supervised/tests/test_label_propagation.py

@@ -122,15 +122,6 @@ def test_label_propagation_closed_form(global_dtype):
     assert_allclose(expected, clf.label_distributions_, atol=1e-4)
 
 
-@pytest.mark.parametrize("alpha", [-0.1, 0, 1, 1.1, None])
-def test_valid_alpha(global_dtype, alpha):
-    n_classes = 2
-    X, y = make_classification(n_classes=n_classes, n_samples=200, random_state=0)
-    X = X.astype(global_dtype)
-    with pytest.raises(ValueError):
-        label_propagation.LabelSpreading(alpha=alpha).fit(X, y)
-
-
 def test_convergence_speed():
     # This is a non-regression test for #5774
     X = np.array([[1.0, 0.0], [0.0, 1.0], [1.0, 2.5]])

sklearn/tests/test_common.py

@@ -469,8 +469,6 @@ def test_estimators_do_not_raise_errors_in_init_or_set_params(Estimator):
     "DictionaryLearning",
     "HashingVectorizer",
     "IterativeImputer",
-    "LabelPropagation",
-    "LabelSpreading",
     "LatentDirichletAllocation",
     "MiniBatchDictionaryLearning",
     "MultiTaskElasticNet",