[FIX] Predictions widget: handle similar but different domains

Orange/widgets/evaluate/owpredictions.py

@@ -316,6 +316,11 @@ def _update_predictions_model(self):
             for p in slots:
                 values, prob = p.results
                 if self.class_var.is_discrete:
+                    # if values were added to class_var between building the
+                    # model and predicting, add zeros for new class values,
+                    # which are always at the end
+                    prob = numpy.c_[prob,
+                                    numpy.zeros((prob.shape[0], len(class_var.values) - prob.shape[1]))]
                     values = [Value(class_var, v) for v in values]
                 results.append((values, prob))
             results = list(zip(*(zip(*res) for res in results)))

Orange/widgets/evaluate/tests/test_owpredictions.py

@@ -1,13 +1,15 @@
 """Tests for OWPredictions"""
-
+import io
 import numpy as np
 
+from Orange.data.io import TabReader
 from Orange.widgets.tests.base import WidgetTest
 from Orange.widgets.evaluate.owpredictions import OWPredictions
 
-from Orange.data import Table, Domain
-from Orange.classification import MajorityLearner
+from Orange.data import Table, Domain, Variable
+from Orange.modelling import ConstantLearner, TreeLearner
 from Orange.evaluation import Results
+from Orange.widgets.tests.utils import excepthook_catch
 
 
 class TestOWPredictions(WidgetTest):
@@ -27,7 +29,7 @@ def test_nan_target_input(self):
         yvec, _ = data.get_column_view(data.domain.class_var)
         nanmask = np.isnan(yvec)
         self.send_signal("Data", data)
-        self.send_signal("Predictors", MajorityLearner()(data), 1)
+        self.send_signal("Predictors", ConstantLearner()(data), 1)
         pred = self.get_output("Predictions", )
         self.assertIsInstance(pred, Table)
         np.testing.assert_array_equal(
@@ -50,8 +52,8 @@ def test_mismatching_targets(self):
         error = self.widget.Error
 
         titanic = Table("titanic")
-        majority_titanic = MajorityLearner()(titanic)
-        majority_iris = MajorityLearner()(self.iris)
+        majority_titanic = ConstantLearner()(titanic)
+        majority_iris = ConstantLearner()(self.iris)
 
         self.send_signal("Data", self.iris)
         self.send_signal("Predictors", majority_iris, 1)
@@ -101,14 +103,14 @@ def test_no_class_on_test(self):
         error = self.widget.Error
 
         titanic = Table("titanic")
-        majority_titanic = MajorityLearner()(titanic)
-        majority_iris = MajorityLearner()(self.iris)
+        majority_titanic = ConstantLearner()(titanic)
+        majority_iris = ConstantLearner()(self.iris)
 
         no_class = Table(Domain(titanic.domain.attributes, None), titanic)
         self.send_signal("Predictors", majority_titanic, 1)
         self.send_signal("Data", no_class)
         out = self.get_output("Predictions")
-        np.testing.assert_allclose(out.get_column_view("majority")[0], 0)
+        np.testing.assert_allclose(out.get_column_view("constant")[0], 0)
 
         self.send_signal("Predictors", majority_iris, 2)
         self.assertTrue(error.predictors_target_mismatch.is_shown())
@@ -118,4 +120,53 @@ def test_no_class_on_test(self):
         self.send_signal("Predictors", None, 2)
         self.send_signal("Data", titanic)
         out = self.get_output("Predictions")
-        np.testing.assert_allclose(out.get_column_view("majority")[0], 0)
+        np.testing.assert_allclose(out.get_column_view("constant")[0], 0)
+
+    def test_bad_data(self):
+        """
+        Firstly it creates predictions with TreeLearner. Then sends predictions and
+        different data with different domain to Predictions widget. Those different
+        data and domain are similar to original data and domain but they have three
+        different target values instead of two.
+        GH-2129
+        """
+        Variable._clear_all_caches()
+
+        filestr1 = """\
+        age\tsex\tsurvived
+        d\td\td
+        \t\tclass
+        adult\tmale\tyes
+        adult\tfemale\tno
+        child\tmale\tyes
+        child\tfemale\tyes
+        """
+        file1 = io.StringIO(filestr1)
+        table = TabReader(file1).read()
+        learner = TreeLearner()
+        tree = learner(table)
+
+        filestr2 = """\
+        age\tsex\tsurvived
+        d\td\td
+        \t\tclass
+        adult\tmale\tyes
+        adult\tfemale\tno
+        child\tmale\tyes
+        child\tfemale\tunknown
+        """
+        file2 = io.StringIO(filestr2)
+        bad_table = TabReader(file2).read()
+
+        self.send_signal("Predictors", tree, 1)
+
+        with excepthook_catch():
+            self.send_signal("Data", bad_table)
+
+        Variable._clear_all_caches()  # so that test excepting standard titanic work
+
+    def test_continuous_class(self):
+        data = Table("housing")
+        cl_data = ConstantLearner()(data)
+        self.send_signal("Predictors", cl_data, 1)
+        self.send_signal("Data", data)