fix: pearson changes inputs (#2765)

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Nicki Skafte Detlefsen <[email protected]>
(cherry picked from commit f3b6278)

CHANGELOG.md

@@ -16,7 +16,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
--
+- Fixed for Pearson changes inputs ([#2765](https://github.com/Lightning-AI/torchmetrics/pull/2765))
 
 ## [1.4.2] - 2022-09-12
 

src/torchmetrics/functional/image/rmse_sw.py

@@ -104,7 +104,8 @@ def _rmse_sw_compute(
     """
     rmse = rmse_val_sum / total_images if rmse_val_sum is not None else None
     if rmse_map is not None:
-        rmse_map /= total_images
+        # prevent overwrite the inputs
+        rmse_map = rmse_map / total_images
     return rmse, rmse_map
 
 

src/torchmetrics/functional/regression/concordance.py

@@ -27,6 +27,8 @@ def _concordance_corrcoef_compute(
 ) -> Tensor:
     """Compute the final concordance correlation coefficient based on accumulated statistics."""
     pearson = _pearson_corrcoef_compute(var_x, var_y, corr_xy, nb)
+    var_x = var_x / (nb - 1)
+    var_y = var_y / (nb - 1)
     return 2.0 * pearson * var_x.sqrt() * var_y.sqrt() / (var_x + var_y + (mean_x - mean_y) ** 2)
 
 

src/torchmetrics/functional/regression/pearson.py

@@ -92,9 +92,10 @@ def _pearson_corrcoef_compute(
         nb: number of observations
 
     """
-    var_x /= nb - 1
-    var_y /= nb - 1
-    corr_xy /= nb - 1
+    # prevent overwrite the inputs
+    var_x = var_x / (nb - 1)
+    var_y = var_y / (nb - 1)
+    corr_xy = corr_xy / (nb - 1)
     # if var_x, var_y is float16 and on cpu, make it bfloat16 as sqrt is not supported for float16
     # on cpu, remove this after https://github.com/pytorch/pytorch/issues/54774 is fixed
     if var_x.dtype == torch.float16 and var_x.device == torch.device("cpu"):

tests/unittests/regression/test_pearson.py

@@ -164,3 +164,25 @@ def test_single_sample_update():
     metric(torch.tensor([7.0]), torch.tensor([8.0]))
     res2 = metric.compute()
     assert torch.allclose(res1, res2)
+
+
+def test_overwrite_reference_inputs():
+    """Test that the normalizations does not overwrite inputs.
+
+    Variables var_x, var_y, corr_xy are references to the object variables and get incorrectly scaled down such that
+    when you update again and compute you get very wrong values.
+
+    """
+    y = torch.randn(100)
+    y_pred = y + torch.randn(y.shape) / 5
+    # Initialize Pearson correlation coefficient metric
+    pearson = PearsonCorrCoef()
+    # Compute the Pearson correlation coefficient
+    correlation = pearson(y, y_pred)
+
+    pearson = PearsonCorrCoef()
+    for lower, upper in [(0, 33), (33, 66), (66, 99), (99, 100)]:
+        pearson.update(torch.tensor(y[lower:upper]), torch.tensor(y_pred[lower:upper]))
+        pearson.compute()
+
+    assert torch.isclose(pearson.compute(), correlation)