Fixed default value for mdmc_average in Accuracy (#1036)

* Fixed default value for mdmc_average in Accuracy as per documentation
* fix tests
* changelog
* typo

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Skaftenicki <skaftenicki@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

CHANGELOG.md

@@ -75,6 +75,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Fixed case where `KLDivergence` could output `Nan` ([#1030](https://github.com/PyTorchLightning/metrics/pull/1030))
 
 
+- Fixed default value for `mdmc_average` in `Accuracy` ([#1036](https://github.com/PyTorchLightning/metrics/pull/1036))
+
+
 ## [0.8.2] - 2022-05-06
 
 

tests/classification/test_accuracy.py

@@ -58,44 +58,44 @@ def _sk_accuracy(preds, target, subset_accuracy):
 
 
 @pytest.mark.parametrize(
- "preds, target, subset_accuracy",
+ "preds, target, subset_accuracy, mdmc_average",
  [
- (_input_binary_logits.preds, _input_binary_logits.target, False),
- (_input_binary_prob.preds, _input_binary_prob.target, False),
- (_input_binary.preds, _input_binary.target, False),
- (_input_mlb_prob.preds, _input_mlb_prob.target, True),
- (_input_mlb_logits.preds, _input_mlb_logits.target, False),
- (_input_mlb_prob.preds, _input_mlb_prob.target, False),
- (_input_mlb.preds, _input_mlb.target, True),
- (_input_mlb.preds, _input_mlb.target, False),
- (_input_mcls_prob.preds, _input_mcls_prob.target, False),
- (_input_mcls_logits.preds, _input_mcls_logits.target, False),
- (_input_mcls.preds, _input_mcls.target, False),
- (_input_mdmc_prob.preds, _input_mdmc_prob.target, False),
- (_input_mdmc_prob.preds, _input_mdmc_prob.target, True),
- (_input_mdmc.preds, _input_mdmc.target, False),
- (_input_mdmc.preds, _input_mdmc.target, True),
- (_input_mlmd_prob.preds, _input_mlmd_prob.target, True),
- (_input_mlmd_prob.preds, _input_mlmd_prob.target, False),
- (_input_mlmd.preds, _input_mlmd.target, True),
- (_input_mlmd.preds, _input_mlmd.target, False),
+ (_input_binary_logits.preds, _input_binary_logits.target, False, None),
+ (_input_binary_prob.preds, _input_binary_prob.target, False, None),
+ (_input_binary.preds, _input_binary.target, False, None),
+ (_input_mlb_prob.preds, _input_mlb_prob.target, True, None),
+ (_input_mlb_logits.preds, _input_mlb_logits.target, False, None),
+ (_input_mlb_prob.preds, _input_mlb_prob.target, False, None),
+ (_input_mlb.preds, _input_mlb.target, True, None),
+ (_input_mlb.preds, _input_mlb.target, False, "global"),
+ (_input_mcls_prob.preds, _input_mcls_prob.target, False, None),
+ (_input_mcls_logits.preds, _input_mcls_logits.target, False, None),
+ (_input_mcls.preds, _input_mcls.target, False, None),
+ (_input_mdmc_prob.preds, _input_mdmc_prob.target, False, "global"),
+ (_input_mdmc_prob.preds, _input_mdmc_prob.target, True, None),
+ (_input_mdmc.preds, _input_mdmc.target, False, "global"),
+ (_input_mdmc.preds, _input_mdmc.target, True, None),
+ (_input_mlmd_prob.preds, _input_mlmd_prob.target, True, None),
+ (_input_mlmd_prob.preds, _input_mlmd_prob.target, False, None),
+ (_input_mlmd.preds, _input_mlmd.target, True, None),
+ (_input_mlmd.preds, _input_mlmd.target, False, "global"),
  ],
 )
 class TestAccuracies(MetricTester):
  @pytest.mark.parametrize("ddp", [False, True])
  @pytest.mark.parametrize("dist_sync_on_step", [False, True])
- def test_accuracy_class(self, ddp, dist_sync_on_step, preds, target, subset_accuracy):
+ def test_accuracy_class(self, ddp, dist_sync_on_step, preds, target, subset_accuracy, mdmc_average):
  self.run_class_metric_test(
  ddp=ddp,
  preds=preds,
  target=target,
  metric_class=Accuracy,
  sk_metric=partial(_sk_accuracy, subset_accuracy=subset_accuracy),
  dist_sync_on_step=dist_sync_on_step,
- metric_args={"threshold": THRESHOLD, "subset_accuracy": subset_accuracy},
+ metric_args={"threshold": THRESHOLD, "subset_accuracy": subset_accuracy, "mdmc_average": mdmc_average},
  )
 
- def test_accuracy_fn(self, preds, target, subset_accuracy):
+ def test_accuracy_fn(self, preds, target, subset_accuracy, mdmc_average):
  self.run_functional_metric_test(
  preds,
  target,
@@ -104,13 +104,13 @@ def test_accuracy_fn(self, preds, target, subset_accuracy):
  metric_args={"threshold": THRESHOLD, "subset_accuracy": subset_accuracy},
  )
 
- def test_accuracy_differentiability(self, preds, target, subset_accuracy):
+ def test_accuracy_differentiability(self, preds, target, subset_accuracy, mdmc_average):
  self.run_differentiability_test(
  preds=preds,
  target=target,
  metric_module=Accuracy,
  metric_functional=accuracy,
- metric_args={"threshold": THRESHOLD, "subset_accuracy": subset_accuracy},
+ metric_args={"threshold": THRESHOLD, "subset_accuracy": subset_accuracy, "mdmc_average": mdmc_average},
  )
 
 
@@ -180,7 +180,7 @@ def test_accuracy_differentiability(self, preds, target, subset_accuracy):
  ],
 )
 def test_topk_accuracy(preds, target, exp_result, k, subset_accuracy):
- topk = Accuracy(top_k=k, subset_accuracy=subset_accuracy)
+ topk = Accuracy(top_k=k, subset_accuracy=subset_accuracy, mdmc_average="global")
 
  for batch in range(preds.shape[0]):
  topk(preds[batch], target[batch])

torchmetrics/classification/accuracy.py

@@ -165,7 +165,7 @@ def __init__(
  threshold: float = 0.5,
  num_classes: Optional[int] = None,
  average: str = "micro",
- mdmc_average: Optional[str] = "global",
+ mdmc_average: Optional[str] = None,
  ignore_index: Optional[int] = None,
  top_k: Optional[int] = None,
  multiclass: Optional[bool] = None,