feat: add ~variance~ standard deviation selector

docs/api.rst

@@ -61,6 +61,7 @@ Selection of the desired metric result from multiple poses.
     OracleSelector
     TopSelector
     RandomSelector
+    DeviationAggregator
 
 Analysis functions
 ------------------

src/peppr/selector.py

@@ -5,6 +5,7 @@
     "OracleSelector",
     "TopSelector",
     "RandomSelector",
+    "DeviationAggregator",
 ]
 
 from abc import ABC, abstractmethod
@@ -174,3 +175,17 @@ def select(self, values: np.ndarray, smaller_is_better: bool) -> float:
             return np.nanmin(top_values)
         else:
             return np.nanmax(top_values)
+
+
+class DeviationAggregator(Selector):
+    """
+    Aggregator that computes the standard deviation of the values. This can be
+    used to assess the consistency of accuracy of a set of predicted poses.
+    """
+
+    @property
+    def name(self) -> str:
+        return "stdev"
+
+    def select(self, values: np.ndarray, smaller_is_better: bool) -> float:
+        return np.nanstd(values)

tests/test_selectors.py

@@ -72,3 +72,17 @@ def test_random_selector():
     ]
 
     assert np.isclose(np.mean(selected_values), 9, rtol=0.5)
+
+
+def test_deviation_aggregator():
+    """
+    This test verifies that the DeviationAggregator returns the expected
+    value of standard deviation for a given set of values.
+    """
+    selector = peppr.DeviationAggregator()
+    values = np.linspace(0, 10, 10 + 1)
+    expected_std = np.std(values)
+
+    selected_value = selector.select(values, smaller_is_better=False)
+
+    assert np.isclose(selected_value, expected_std)