DOC Clarify what can be passed to the plotting utilities (scikit-learn#16893)

Author: justmarkham

sklearn/metrics/_plot/confusion_matrix.py

@@ -137,7 +137,8 @@ def plot_confusion_matrix(estimator, X, y_true, labels=None,
     Parameters
     ----------
     estimator : estimator instance
-        Trained classifier.
+        Fitted classifier or a fitted :class:`~sklearn.pipeline.Pipeline`
+        in which the last estimator is a classifier.
 
     X : {array-like, sparse matrix} of shape (n_samples, n_features)
         Input values.

sklearn/metrics/_plot/precision_recall_curve.py

@@ -106,7 +106,8 @@ def plot_precision_recall_curve(estimator, X, y,
     Parameters
     ----------
     estimator : estimator instance
-        Trained classifier.
+        Fitted classifier or a fitted :class:`~sklearn.pipeline.Pipeline`
+        in which the last estimator is a classifier.
 
     X : {array-like, sparse matrix} of shape (n_samples, n_features)
         Input values.

sklearn/metrics/_plot/roc_curve.py

@@ -115,7 +115,8 @@ def plot_roc_curve(estimator, X, y, sample_weight=None,
     Parameters
     ----------
     estimator : estimator instance
-        Trained classifier.
+        Fitted classifier or a fitted :class:`~sklearn.pipeline.Pipeline`
+        in which the last estimator is a classifier.
 
     X : {array-like, sparse matrix} of shape (n_samples, n_features)
         Input values.