fix typo in specificity_at_sensitivity (#2199)

Co-authored-by: Jirka Borovec <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Nicki Skafte Detlefsen <[email protected]>
Lightning-AI · Nov 29, 2023 · b7f74c2 · b7f74c2
1 parent b07b2a5
commit b7f74c2
Show file tree

Hide file tree

Showing 4 changed files with 60 additions and 21 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -38,6 +38,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Changed minimum supported Pytorch version from 1.8 to 1.10 ([#2145](https://github.com/Lightning-AI/torchmetrics/pull/2145))
 
 
+- Changed x-/y-axis order for `PrecisionRecallCurve` to be consistent with scikit-learn ([#2183](https://github.com/Lightning-AI/torchmetrics/pull/2183))
+
+
 - Use arange and repeat for deterministic bincount ([#2184](https://github.com/Lightning-AI/torchmetrics/pull/2184))
 
 
@@ -46,7 +49,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Deprecated `metric._update_called` ([#2141](https://github.com/Lightning-AI/torchmetrics/pull/2141))
 
 
-- Changed x-/y-axis order for `PrecisionRecallCurve` to be consistent with scikit-learn ([#2183](https://github.com/Lightning-AI/torchmetrics/pull/2183))
+- Deprecated `specicity_at_sensitivity` in favour of `specificity_at_sensitivity` ([#2199](https://github.com/Lightning-AI/torchmetrics/pull/2199))
 
 
 ### Removed

diff --git a/src/torchmetrics/classification/__init__.py b/src/torchmetrics/classification/__init__.py
@@ -117,18 +117,6 @@
 )
 
 __all__ = [
-    "BinaryConfusionMatrix",
-    "ConfusionMatrix",
-    "MulticlassConfusionMatrix",
-    "MultilabelConfusionMatrix",
-    "PrecisionRecallCurve",
-    "BinaryPrecisionRecallCurve",
-    "MulticlassPrecisionRecallCurve",
-    "MultilabelPrecisionRecallCurve",
-    "BinaryStatScores",
-    "MulticlassStatScores",
-    "MultilabelStatScores",
-    "StatScores",
     "Accuracy",
     "BinaryAccuracy",
     "MulticlassAccuracy",
@@ -147,6 +135,10 @@
     "BinaryCohenKappa",
     "CohenKappa",
     "MulticlassCohenKappa",
+    "BinaryConfusionMatrix",
+    "ConfusionMatrix",
+    "MulticlassConfusionMatrix",
+    "MultilabelConfusionMatrix",
     "Dice",
     "ExactMatch",
     "MulticlassExactMatch",
@@ -184,29 +176,34 @@
     "MultilabelRecall",
     "Precision",
     "Recall",
+    "BinaryPrecisionRecallCurve",
+    "MulticlassPrecisionRecallCurve",
+    "MultilabelPrecisionRecallCurve",
+    "PrecisionRecallCurve",
     "MultilabelCoverageError",
     "MultilabelRankingAveragePrecision",
     "MultilabelRankingLoss",
+    "RecallAtFixedPrecision",
     "BinaryRecallAtFixedPrecision",
     "MulticlassRecallAtFixedPrecision",
     "MultilabelRecallAtFixedPrecision",
-    "ROC",
     "BinaryROC",
     "MulticlassROC",
     "MultilabelROC",
+    "ROC",
     "BinarySpecificity",
     "MulticlassSpecificity",
     "MultilabelSpecificity",
     "Specificity",
     "BinarySpecificityAtSensitivity",
     "MulticlassSpecificityAtSensitivity",
     "MultilabelSpecificityAtSensitivity",
-    "BinaryPrecisionAtFixedRecall",
     "SpecificityAtSensitivity",
-    "MulticlassPrecisionAtFixedRecall",
-    "MultilabelPrecisionAtFixedRecall",
+    "BinaryStatScores",
+    "MulticlassStatScores",
+    "MultilabelStatScores",
+    "StatScores",
     "PrecisionAtFixedRecall",
-    "RecallAtFixedPrecision",
     "BinaryPrecisionAtFixedRecall",
     "MulticlassPrecisionAtFixedRecall",
     "MultilabelPrecisionAtFixedRecall",

diff --git a/src/torchmetrics/functional/classification/__init__.py b/src/torchmetrics/functional/classification/__init__.py
@@ -120,6 +120,7 @@
     multiclass_specificity_at_sensitivity,
     multilabel_specificity_at_sensitivity,
     specicity_at_sensitivity,
+    specificity_at_sensitivity,
 )
 from torchmetrics.functional.classification.stat_scores import (
     binary_stat_scores,
@@ -165,8 +166,6 @@
     "multilabel_fbeta_score",
     "binary_fairness",
     "binary_groups_stat_rates",
-    "demographic_parity",
-    "equal_opportunity",
     "binary_hamming_distance",
     "hamming_distance",
     "multiclass_hamming_distance",
@@ -212,11 +211,14 @@
     "multiclass_specificity_at_sensitivity",
     "multilabel_specificity_at_sensitivity",
     "specicity_at_sensitivity",
+    "specificity_at_sensitivity",
     "binary_stat_scores",
     "multiclass_stat_scores",
     "multilabel_stat_scores",
     "stat_scores",
     "binary_precision_at_fixed_recall",
     "multilabel_precision_at_fixed_recall",
     "multiclass_precision_at_fixed_recall",
+    "demographic_parity",
+    "equal_opportunity",
 ]
diff --git a/src/torchmetrics/functional/classification/specificity_sensitivity.py b/src/torchmetrics/functional/classification/specificity_sensitivity.py
@@ -11,6 +11,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import warnings
 from typing import List, Optional, Tuple, Union
 
 import torch
@@ -414,7 +415,43 @@ def specicity_at_sensitivity(
     ignore_index: Optional[int] = None,
     validate_args: bool = True,
 ) -> Union[Tensor, Tuple[Tensor, Tensor, Tensor], Tuple[List[Tensor], List[Tensor], List[Tensor]]]:
-    r"""Compute the highest possible specicity value given the minimum sensitivity thresholds provided.
+    r"""Compute the highest possible specificity value given the minimum sensitivity thresholds provided.
+
+    .. warning::
+        This function was deprecated in v1.3.0 of Torchmetrics and will be removed in v2.0.0.
+        Use `specificity_at_sensitivity` instead.
+
+    """
+    warnings.warn(
+        "This method has will be removed in 2.0.0. Use `specificity_at_sensitivity` instead.",
+        DeprecationWarning,
+        stacklevel=1,
+    )
+    return specificity_at_sensitivity(
+        preds=preds,
+        target=target,
+        task=task,
+        min_sensitivity=min_sensitivity,
+        thresholds=thresholds,
+        num_classes=num_classes,
+        num_labels=num_labels,
+        ignore_index=ignore_index,
+        validate_args=validate_args,
+    )
+
+
+def specificity_at_sensitivity(
+    preds: Tensor,
+    target: Tensor,
+    task: Literal["binary", "multiclass", "multilabel"],
+    min_sensitivity: float,
+    thresholds: Optional[Union[int, List[float], Tensor]] = None,
+    num_classes: Optional[int] = None,
+    num_labels: Optional[int] = None,
+    ignore_index: Optional[int] = None,
+    validate_args: bool = True,
+) -> Union[Tensor, Tuple[Tensor, Tensor, Tensor], Tuple[List[Tensor], List[Tensor], List[Tensor]]]:
+    r"""Compute the highest possible specificity value given the minimum sensitivity thresholds provided.
 
     This is done by first calculating the Receiver Operating Characteristic (ROC) curve for different thresholds and
     the find the specificity for a given sensitivity level.