Merge pull request #5514 from janezd/hierarchical-annotate-vars

[ENH] Hierarchical Clustering: Annotate variables with clusters

Author: markotoplak

Orange/widgets/unsupervised/owhierarchicalclustering.py

@@ -610,16 +610,29 @@ def commit(self):
                 selected_data.domain = Domain(
                     attrs, classes, metas + (clust_var, ))
 
+            annotated_data = create_annotated_table(data, selected_indices)
+
         elif isinstance(items, Orange.data.Table) and self.matrix.axis == 0:
             # Select columns
+            attrs = []
+            for clust, indices in chain(enumerate(maps, start=1),
+                                        [(0, unselected_indices)]):
+                for i in indices:
+                    attr = items.domain[i].copy()
+                    attr.attributes["cluster"] = clust
+                    attrs.append(attr)
             domain = Orange.data.Domain(
-                [items.domain[i] for i in selected_indices],
+                # len(unselected_indices) can be 0
+                attrs[:len(attrs) - len(unselected_indices)],
                 items.domain.class_vars, items.domain.metas)
             selected_data = items.from_table(domain, items)
-            data = None
+
+            domain = Orange.data.Domain(
+                attrs,
+                items.domain.class_vars, items.domain.metas)
+            annotated_data = items.from_table(domain, items)
 
         self.Outputs.selected_data.send(selected_data)
-        annotated_data = create_annotated_table(data, selected_indices)
         self.Outputs.annotated_data.send(annotated_data)
 
     def eventFilter(self, obj, event):

Orange/widgets/unsupervised/tests/test_owhierarchicalclustering.py

@@ -26,6 +26,8 @@ def setUpClass(cls):
         cls.signal_data = cls.distances
         cls.same_input_output_domain = False
 
+        cls.distances_cols = Euclidean(cls.data, axis=0)
+
     def setUp(self):
         self.widget = self.create_widget(OWHierarchicalClustering)
 
@@ -35,6 +37,11 @@ def _select_data(self):
         self.widget.dendrogram.set_selected_items([cluster])
         return [14, 15, 32, 33]
 
+    def _select_data_columns(self):
+        items = self.widget.dendrogram._items
+        cluster = items[sorted(list(items.keys()))[5]]
+        self.widget.dendrogram.set_selected_items([cluster])
+
     def _compare_selected_annotated_domains(self, selected, annotated):
         self.assertEqual(annotated.domain.variables,
                          selected.domain.variables)
@@ -170,3 +177,17 @@ def test_restore_state(self):
         self.send_signal(w.Inputs.distances, self.distances, widget=w)
         ids_2 = self.get_output(w.Outputs.selected_data, widget=w).ids
         self.assertSequenceEqual(list(ids_1), list(ids_2))
+
+    def test_column_distances(self):
+        self.send_signal(self.widget.Inputs.distances, self.distances_cols)
+        self._select_data_columns()
+        o = self.get_output(self.widget.Outputs.annotated_data)
+        annotated = [(a.name, a.attributes['cluster']) for a in o.domain.attributes]
+        self.assertEqual(annotated, [('sepal width', 1), ('petal length', 1),
+                                     ('sepal length', 0), ('petal width', 0)])
+
+        self.widget.selection_box.buttons[2].click()  # top N
+        o = self.get_output(self.widget.Outputs.annotated_data)
+        annotated = [(a.name, a.attributes['cluster']) for a in o.domain.attributes]
+        self.assertEqual(annotated, [('sepal length', 1), ('petal width', 2),
+                                     ('sepal width', 3), ('petal length', 3)])