Merge pull request #20 from jeethu/jeethu/clobber-prevention

Make numerai_corr a pure function

numerai_tools/scoring.py

@@ -403,7 +403,7 @@ def numerai_corr(
     Returns:
         pd.Series - the resulting correlation scores for each column in predictions
     """
-    targets -= targets.mean()
+    targets = targets - targets.mean()
     targets, predictions = filter_sort_index(
         targets, predictions, max_filtered_index_ratio
     )

tests/test_scoring.py

@@ -5,6 +5,7 @@
 
 from numerai_tools.scoring import (
     correlation,
+    numerai_corr,
     tie_broken_rank_correlation,
     spearman_correlation,
     pearson_correlation,
@@ -198,3 +199,12 @@ def test_neutralize(self):
                 [0, 0, 0, 0, 0],
             ],
         ).all()
+
+    def test_numerai_corr_doesnt_clobber_targets(self):
+        s = [x/4 for x in range(5)]
+        df = pd.DataFrame({
+            "target": s,
+            "prediction": reversed(s)
+        })
+        numerai_corr(df[["prediction"]], df["target"])
+        assert pd.Series(s).equals(df["target"]), f"{s} != {list(df['target'].values)}"