[ENH] Box Plot: Add box for missing group values

Orange/statistics/contingency.py

@@ -225,6 +225,18 @@ def from_data(self, data, col_variable, row_variable=None):
                 "Fallback method for computation of contingencies is not implemented yet"
             )
 
+    @property
+    def array_with_unknowns(self):
+        """
+        This function returns the list of all items returned by __getitem__
+        with adding a row of row_unknowns together with values.
+        """
+        # pylint: disable=unnecessary-comprehension
+        other_rows = [x for x in self]
+        ind = self.row_unknowns > 0
+        unknown_rows = np.vstack((self.values[ind], self.row_unknowns[ind]))
+        return other_rows + [unknown_rows]
+
     def __eq__(self, other):
         return (
             np.array_equal(self.values, other.values) and

Orange/tests/test_contingency.py

@@ -165,6 +165,32 @@ def test_continuous_missing(self):
              3., 4., 2., 1., 1., 1., 1.])
         self.assertEqual(cont.unknowns, 1)
 
+    @staticmethod
+    def test_continuous_array_with_unknowns():
+        """
+        Test array_with_unknowns function
+        """
+        d = data.Table("iris")
+        d.Y[:50] = np.zeros(50) * float("nan")
+        cont = contingency.Continuous(d, "sepal width")
+        correct_row_unknowns = [0., 0., 1., 0., 0., 0., 0., 0., 1., 6., 5., 5.,
+                                2., 9., 6., 2., 3., 4., 2., 1., 1., 1., 1.]
+        correct_row_unknowns_no_zero = [
+            c for c in correct_row_unknowns if c > 0]
+        correct_values_no_zero = [
+            v for v, c in zip(cont.values, correct_row_unknowns) if c > 0]
+
+        np.testing.assert_almost_equal(cont.row_unknowns, correct_row_unknowns)
+        arr_unknowns = cont.array_with_unknowns
+        np.testing.assert_almost_equal(
+            arr_unknowns[-1][1], correct_row_unknowns_no_zero)
+        np.testing.assert_almost_equal(
+            arr_unknowns[-1][0], correct_values_no_zero)
+
+        # check if other match to what we get with __getitem__
+        for v1, v2 in zip(arr_unknowns[:-1], cont):
+            np.testing.assert_almost_equal(v1, v2)
+
     def test_mixedtype_metas(self):
         import Orange
         zoo = Orange.data.Table("zoo")
@@ -212,7 +238,6 @@ def _construct_sparse():
         Y = np.array([[1, 2, 1, 0, 0]]).T
         return data.Table.from_numpy(domain, X, Y)
 
-
     def test_sparse(self):
         d = self._construct_sparse()
         cont = contingency.Discrete(d, 5)

Orange/widgets/visualize/owboxplot.py

@@ -494,20 +494,21 @@ def compute_box_data(self):
             self.dist = []
             self.conts = contingency.get_contingency(
                 dataset, attr, self.group_var)
+            group_var_labels = self.group_var.values + [
+                f"missing '{self.group_var.name}'"]
             if self.is_continuous:
                 stats, label_texts = [], []
-                for i, cont in enumerate(self.conts):
+                for i, cont in enumerate(self.conts.array_with_unknowns):
                     if np.sum(cont[1]):
                         stats.append(BoxData(cont, attr, i, self.group_var))
-                        label_texts.append(self.group_var.values[i])
+                        label_texts.append(group_var_labels[i])
                 self.stats = stats
                 self.label_txts_all = label_texts
             else:
-                self.label_txts_all = \
-                    [v for v, c in zip(
-                        self.group_var.values + ["Missing values"],
-                        self.conts.array_with_unknowns)
-                     if np.sum(c) > 0]
+                self.label_txts_all = [
+                    v for v, c in zip(
+                        group_var_labels, self.conts.array_with_unknowns)
+                    if np.sum(c) > 0]
         else:
             self.dist = distribution.get_distribution(dataset, attr)
             self.conts = []
@@ -1071,7 +1072,7 @@ def strudel(self, dist, group_val_index=None):
                 cond.append(FilterDiscrete(self.group_var, [group_val_index]))
             box.append(FilterGraphicsRectItem(cond, 0, -10, 1, 10))
         cum = 0
-        values = attr.values + ["Missing values"]
+        values = attr.values + [f"missing '{attr.name}'"]
         colors = np.vstack((attr.colors, [128, 128, 128]))
         for i, v in enumerate(dist):
             if v < 1e-6: