Merge pull request #137 from Pennycook/code-divergence

Fix definition and implementation of divergence

Author: Pennycook

codebasin/report.py

@@ -77,7 +77,7 @@ def divergence(setmap):
         npairs += 1
 
     if npairs == 0:
-        return 0
+        return float("nan")
     return d / float(npairs)
 
 

docs/source/specialization.rst

@@ -57,7 +57,7 @@ pairwise distances:
 
 .. math::
     CD(a, p, H) = \binom{|H|}{2}^{-1}
-                  \sum_{\{i, j\} \in H \times H} {d_{i, j}(a, p)}
+                  \sum_{\{i, j\} \in \binom{H}{2}} {d_{i, j}(a, p)}
 
 where :math:`d_{i, j}(a, p)` represents the distance between the source
 code required by platforms :math:`i` and :math:`j` for application

tests/metrics/__init__.py

@@ -0,0 +1,49 @@
+# Copyright (C) 2019-2024 Intel Corporation
+# SPDX-License-Identifier: BSD-3-Clause
+
+import logging
+import math
+import unittest
+import warnings
+
+from codebasin.report import divergence
+
+
+class TestDivergence(unittest.TestCase):
+    """
+    Test computation of code divergence.
+    """
+
+    def setUp(self):
+        logging.disable()
+        warnings.simplefilter("ignore", ResourceWarning)
+
+    def test_divergence(self):
+        """Check divergence computation for simple setmap."""
+        setmap = {
+            frozenset(["A"]): 1,
+            frozenset(["B"]): 2,
+            frozenset(["A", "B"]): 3,
+            frozenset([]): 4,
+        }
+        intersection = 3
+        union = 1 + 2 + 3
+
+        expected_divergence = intersection / union
+        self.assertEqual(divergence(setmap), expected_divergence)
+
+    def test_null_divergence(self):
+        """Check divergence computation for null cases."""
+        setmap = {
+            frozenset(""): 0,
+        }
+        self.assertTrue(math.isnan(divergence(setmap)))
+
+        setmap = {
+            frozenset("A"): 1,
+        }
+        self.assertTrue(math.isnan(divergence(setmap)))
+
+
+if __name__ == "__main__":
+    unittest.main()