DOC Improve docs of BaseLabelPropagation.transduction_ (scikit-learn#24985)

scottgigante-immunai · jjerphan · commit 11f9e82db88f · 2023-01-20T15:18:15.000+01:00
diff --git a/sklearn/semi_supervised/_label_propagation.py b/sklearn/semi_supervised/_label_propagation.py
@@ -184,6 +184,10 @@ def predict(self, X):
         y : ndarray of shape (n_samples,)
             Predictions for input data.
         """
+        # Note: since `predict` does not accept semi-supervised labels as input,
+        # `fit(X, y).predict(X) != fit(X, y).transduction_`.
+        # Hence, `fit_predict` is not implemented.
+        # See https://github.com/scikit-learn/scikit-learn/pull/24898
         probas = self.predict_proba(X)
         return self.classes_[np.argmax(probas, axis=1)].ravel()
 
@@ -244,7 +248,7 @@ def fit(self, X, y):
         y : array-like of shape (n_samples,)
             Target class values with unlabeled points marked as -1.
             All unlabeled samples will be transductively assigned labels
-            internally.
+            internally, which are stored in `transduction_`.
 
         Returns
         -------
@@ -371,7 +375,7 @@ class LabelPropagation(BaseLabelPropagation):
         Categorical distribution for each item.
 
     transduction_ : ndarray of shape (n_samples)
-        Label assigned to each item via the transduction.
+        Label assigned to each item during :term:`fit`.
 
     n_features_in_ : int
         Number of features seen during :term:`fit`.
@@ -466,7 +470,7 @@ def fit(self, X, y):
         y : array-like of shape (n_samples,)
             Target class values with unlabeled points marked as -1.
             All unlabeled samples will be transductively assigned labels
-            internally.
+            internally, which are stored in `transduction_`.
 
         Returns
         -------
@@ -531,7 +535,7 @@ class LabelSpreading(BaseLabelPropagation):
         Categorical distribution for each item.
 
     transduction_ : ndarray of shape (n_samples,)
-        Label assigned to each item via the transduction.
+        Label assigned to each item during :term:`fit`.
 
     n_features_in_ : int
         Number of features seen during :term:`fit`.
