util.stats supports weights for variance

Author: markotoplak

Orange/statistics/util.py

@@ -341,31 +341,31 @@ def stats(X, weights=None, compute_variance=False):
     """
     is_numeric = np.issubdtype(X.dtype, np.number)
     is_sparse = sp.issparse(X)
-    weighted = weights is not None and X.dtype != object
-    weights = weights if weighted else None
+
+    if X.size and is_numeric:
+        if compute_variance:
+            means, vars = nan_mean_variance_axis(X, axis=0, weights=weights)
+        else:
+            means = nanmean(X, axis=0, weights=weights)
+            vars = np.zeros(X.shape[1] if X.ndim == 2 else 1)
 
     if X.size and is_numeric and not is_sparse:
         nans = np.isnan(X).sum(axis=0)
         return np.column_stack((
             np.nanmin(X, axis=0),
             np.nanmax(X, axis=0),
-            nanmean(X, axis=0, weights=weights),
-            nanvar(X, axis=0) if compute_variance else \
-                np.zeros(X.shape[1] if X.ndim == 2 else 1),
+            means,
+            vars,
             nans,
             X.shape[0] - nans))
     elif is_sparse and X.size:
-        if compute_variance and weighted:
-            raise NotImplementedError
-
         non_zero = np.bincount(X.nonzero()[1], minlength=X.shape[1])
         X = X.tocsc()
         return np.column_stack((
             nanmin(X, axis=0),
             nanmax(X, axis=0),
-            nanmean(X, axis=0, weights=weights),
-            nanvar(X, axis=0) if compute_variance else \
-                np.zeros(X.shape[1] if X.ndim == 2 else 1),
+            means,
+            vars,
             X.shape[0] - non_zero,
             non_zero))
     else:

Orange/tests/test_statistics.py

@@ -155,8 +155,9 @@ def test_stats_weights_sparse(self):
         np.testing.assert_equal(stats(X, weights), [[0, 2, 1.5, 0, 1, 1],
                                                     [1, 3, 2.5, 0, 0, 2]])
 
-        with self.assertRaises(NotImplementedError):
-            stats(X, weights, compute_variance=True)
+        np.testing.assert_equal(stats(X, weights, compute_variance=True),
+                                [[0, 2, 1.5, 0.75, 1, 1],
+                                 [1, 3, 2.5, 0.75, 0, 2]])
 
     def test_stats_non_numeric(self):
         X = np.array([