Merge pull request #6212 from janezd/preprocess-missing-reprs

[FIX] Preprocess: Fix reporting for remove sparse and impute

Author: VesnaT

Orange/widgets/data/owpreprocess.py

@@ -27,6 +27,7 @@
 from Orange.preprocess import Continuize, ProjectPCA, RemoveNaNRows, \
     ProjectCUR, Scale as _Scale, Randomize as _Randomize, RemoveSparse
 from Orange.widgets import widget, gui
+from Orange.widgets.utils.localization import pl
 from Orange.widgets.settings import Setting
 from Orange.widgets.utils.overlay import OverlayWidget
 from Orange.widgets.utils.sql import check_sql_input
@@ -253,9 +254,9 @@ def createinstance(params):
     def __repr__(self):
         return self.Continuizers[self.__treatment]
 
-class RemoveSparseEditor(BaseEditor):
 
-    options = ["missing", "zeros"]
+class RemoveSparseEditor(BaseEditor):
+    options = ["missing values", "zeros"]
 
     def __init__(self, parent=None, **kwargs):
         super().__init__(parent, **kwargs)
@@ -266,11 +267,9 @@ def __init__(self, parent=None, **kwargs):
         self.setLayout(QVBoxLayout())
 
         self.layout().addWidget(QLabel("Remove features with too many"))
-        options = ["missing values",
-                   "zeros"]
         self.filter_buttons = QButtonGroup(exclusive=True)
         self.filter_buttons.buttonClicked.connect(self.filterByClicked)
-        for idx, option, in enumerate(options):
+        for idx, option, in enumerate(self.options):
             btn = QRadioButton(self, text=option, checked=idx == 0)
             self.filter_buttons.addButton(btn, id=idx)
             self.layout().addWidget(btn)
@@ -354,6 +353,15 @@ def createinstance(params):
             threshold = params.pop('percThresh', 5) / 100
         return RemoveSparse(threshold, filter0)
 
+    def __repr__(self):
+        desc = f"remove features with too many {self.options[self.filter0]}, threshold: "
+        if self.useFixedThreshold:
+            desc += f"{self.fixedThresh} {pl(self.fixedThresh, 'instance')}"
+        else:
+            desc += f"{self.percThresh} %"
+        return desc
+
+
 class ImputeEditor(BaseEditor):
     (NoImputation, Constant, Average,
      Model, Random, DropRows, DropColumns) = 0, 1, 2, 3, 4, 5, 6
@@ -727,6 +735,12 @@ def createinstance(params):
             # further implementations
             raise NotImplementedError
 
+    def __repr__(self):
+        if self.__strategy == self.Fixed:
+            return f"select {self.__k} {pl(self.__k,'feature')}"
+        else:
+            return f"select {self.__p} % features"
+
 
 def index_to_enum(enum, i):
     """Enums, by default, are not int-comparable, so use an ad-hoc mapping of

Orange/widgets/data/tests/test_owpreprocess.py

@@ -1,5 +1,7 @@
 # Test methods with long descriptive names can omit docstrings
 # pylint: disable=missing-docstring,unsubscriptable-object
+import unittest
+
 import numpy as np
 
 from Orange.data import Table
@@ -231,6 +233,13 @@ def test_editor(self):
         self.assertIsInstance(p, fss.SelectRandomFeatures)
         self.assertEqual(p.k, 0.25)
 
+    def test_repr(self):
+        widget = owpreprocess.RandomFeatureSelectEditor()
+        for strategy in (owpreprocess.RandomFeatureSelectEditor.Fixed,
+                         owpreprocess.RandomFeatureSelectEditor.Percentage):
+            widget.setStrategy(strategy)
+            repr(widget)
+
 
 class TestRandomizeEditor(WidgetTest):
     def test_editor(self):
@@ -281,8 +290,8 @@ def test_editor(self):
         self.assertEqual(p.rank, 5)
         self.assertEqual(p.max_error, 0.5)
 
-class TestRemoveSparseEditor(WidgetTest):
 
+class TestRemoveSparseEditor(WidgetTest):
     def test_editor(self):
         widget = owpreprocess.RemoveSparseEditor()
         self.assertEqual(
@@ -304,3 +313,12 @@ def test_editor(self):
         self.assertIsInstance(p, RemoveSparse)
         self.assertEqual(p.threshold, 30)
         self.assertFalse(p.filter0)
+
+    def test_repr(self):
+        widget = owpreprocess.RemoveSparseEditor()
+        for widget.useFixedThreshold in (False, True):
+            repr(widget)
+
+
+if __name__ == "__main__":
+    unittest.main()