LogisticRegressionLearner: do not use MajorityLearner when one class

Author: PrimozGodec

Orange/classification/logistic_regression.py

@@ -1,10 +1,6 @@
-import warnings
-
 import numpy as np
-
 import sklearn.linear_model as skl_linear_model
 
-import Orange
 from Orange.classification import SklLearner, SklModel
 from Orange.preprocess import Normalize
 from Orange.preprocess.score import LearnerScorer
@@ -44,12 +40,3 @@ def __init__(self, penalty="l2", dual=False, tol=0.0001, C=1.0,
                  multi_class='ovr', verbose=0, n_jobs=1, preprocessors=None):
         super().__init__(preprocessors=preprocessors)
         self.params = vars()
-
-    def __call__(self, data):
-        if len(np.unique(data.Y)) > 1:
-            return super().__call__(data)
-        else:
-            warnings.warn("Single class in data, returning Constant Model.")
-            maj = Orange.classification.MajorityLearner()
-            const = maj(data)
-            return const

Orange/tests/test_logistic_regression.py

@@ -119,16 +119,6 @@ def test_predict_on_instance(self):
         probs2 = m(self.zoo[50, :], m.Probs)
         np.testing.assert_almost_equal(probs, probs2[0])
 
-    def test_single_class(self):
-        t = self.iris[60:90]
-        self.assertEqual(len(np.unique(t.Y)), 1)
-        learn = LogisticRegressionLearner()
-        with self.assertWarns(UserWarning):
-            model = learn(t)
-        self.assertEqual(model(t[0]), 1)
-        self.assertTrue(np.all(model(t[0], ret=Model.Probs) == [0, 1, 0]))
-        self.assertTrue(np.all(model(t) == 1))
-
     def test_sklearn_single_class(self):
         t = self.iris[60:90]
         self.assertEqual(len(np.unique(t.Y)), 1)