[FIX] Enable classification tests

Orange/classification/tests/test_base.py

@@ -1,5 +1,3 @@
-import os
-from tempfile import TemporaryDirectory
 import unittest
 
 import numpy as np
@@ -12,17 +10,21 @@ class TestModelMapping(unittest.TestCase):
     @classmethod
     def setUpClass(cls):
         cls.iris = iris = Table("iris")
-        with TemporaryDirectory() as tempdir:
-            tables = []
-            x = np.vstack((iris.X[:50], iris.X[100:]))
-            y = np.hstack((iris.Y[:50], iris.Y[100:]))
-            for i, data in enumerate([iris[50:],
-                                      Table.from_numpy(iris.domain, x, y),
-                                      iris[:100]]):
-
-                name = os.path.join(tempdir, f"no{i}.tab")
-                data.save(name)
-                tables.append(Table(name))
+
+        tables = []
+        ix = iris.X
+        y = np.hstack((np.zeros(50), np.ones(50)))
+        attrs = cls.iris.domain.attributes
+        classes = cls.iris.domain.class_var.values
+        for i, x in enumerate([ix[50:],
+                               np.vstack((ix[:50], ix[100:])),
+                               ix[:100]]):
+            class_var = DiscreteVariable(
+                "iris",
+                values=tuple(n for j, n in enumerate(classes) if j != i))
+            domain = Domain(attrs, class_var)
+            tables.append(Table.from_numpy(domain, x, y))
+        # pylint: disable=unbalanced-tuple-unpacking
         cls.iris0, cls.iris1, cls.iris2 = tables
 
     def test_larger_model(self):

Orange/classification/tests/test_calibration.py

@@ -79,19 +79,22 @@ def test_fit_storage(self, test_on_training, curves_from_results):
         model.domain.class_var.values = ("a", "b")
         data = Table("heart_disease")
         learner = Mock()
-        test_on_training.return_value = res = Mock()
+        test_on_training.return_value = tot = Mock()
+        res = Mock()
         res.models = np.array([[model]])
-        test_on_training.return_value = res
+        tot.return_value = res
 
         thresh_learner = ThresholdLearner(
             base_learner=learner,
             threshold_criterion=ThresholdLearner.OptimizeCA)
         thresh_model = thresh_learner(data)
         self.assertEqual(thresh_model.threshold, 0.15)
-        args, kwargs = test_on_training.call_args
+        args, _ = tot.call_args  # pylint: disable=unpacking-non-sequence
         self.assertEqual(len(args), 2)
         self.assertIs(args[0], data)
         self.assertIs(args[1][0], learner)
+
+        _, kwargs = test_on_training.call_args
         self.assertEqual(len(args[1]), 1)
         self.assertEqual(kwargs, {"store_models": 1})
 
@@ -178,10 +181,11 @@ def test_fit_storage(self, test_on_training, sigmoid_fit):
         model.domain.class_var.is_discrete = True
         model.domain.class_var.values = ("a", "b")
 
-        test_on_training.return_value = res = Mock()
+        test_on_training.return_value = tot = Mock()
+        res = Mock()
         res.models = np.array([[model]])
         res.probabilities = np.arange(20, dtype=float).reshape(1, 5, 4)
-        test_on_training.return_value = res
+        tot.return_value = res
 
         sigmoid_fit.return_value = Mock()
 
@@ -191,11 +195,13 @@ def test_fit_storage(self, test_on_training, sigmoid_fit):
 
         self.assertIs(cal_model.base_model, model)
         self.assertEqual(cal_model.calibrators, [sigmoid_fit.return_value] * 4)
-        args, kwargs = test_on_training.call_args
+        args, _ = tot.call_args  # pylint: disable=unpacking-non-sequence
         self.assertEqual(len(args), 2)
         self.assertIs(args[0], data)
         self.assertIs(args[1][0], learner)
         self.assertEqual(len(args[1]), 1)
+
+        _, kwargs = test_on_training.call_args
         self.assertEqual(kwargs, {"store_models": 1})
 
         for call, cls_probs in zip(sigmoid_fit.call_args_list,