Newmetric: Negative predictive value

CHANGELOG.md

@@ -15,6 +15,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Added `NormalizedRootMeanSquaredError` metric to regression subpackage ([#2442](https://github.com/Lightning-AI/torchmetrics/pull/2442))
 
 
+- Added `NegativePredictiveValue` to classification metrics ([#2433](https://github.com/Lightning-AI/torchmetrics/pull/2433))
+
+
 ### Changed
 
 - Changed naming and input order arguments in `KLDivergence` ([#2800](https://github.com/Lightning-AI/torchmetrics/pull/2800))

docs/source/classification/negative_predictive_value.rst

@@ -0,0 +1,56 @@
+.. customcarditem::
+   :header: Negative Predictive Value
+   :image: https://pl-flash-data.s3.amazonaws.com/assets/thumbnails/tabular_classification.svg
+   :tags: Classification
+
+.. include:: ../links.rst
+
+#########################
+Negative Predictive Value
+#########################
+
+Module Interface
+________________
+
+.. autoclass:: torchmetrics.NegativePredictiveValue
+    :exclude-members: update, compute
+    :special-members: __new__
+
+BinaryNegativePredictiveValue
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.classification.BinaryNegativePredictiveValue
+    :exclude-members: update, compute
+
+MulticlassNegativePredictiveValue
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.classification.MulticlassNegativePredictiveValue
+    :exclude-members: update, compute
+
+MultilabelNegativePredictiveValue
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.classification.MultilabelNegativePredictiveValue
+    :exclude-members: update, compute
+
+
+Functional Interface
+____________________
+
+.. autofunction:: torchmetrics.functional.negative_predictive_value
+
+binary_negative_predictive_value
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.classification.binary_negative_predictive_value
+
+multiclass_negative_predictive_value
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.classification.multiclass_negative_predictive_value
+
+multilabel_negative_predictive_value
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.classification.multilabel_negative_predictive_value

docs/source/links.rst

@@ -176,3 +176,4 @@
 .. _Hausdorff Distance: https://en.wikipedia.org/wiki/Hausdorff_distance
 .. _averaging curve objects: https://scikit-learn.org/stable/auto_examples/model_selection/plot_roc.html
 .. _Procrustes Disparity: https://en.wikipedia.org/wiki/Procrustes_analysis
+.. _Negative Predictive Value: https://en.wikipedia.org/wiki/Positive_and_negative_predictive_values

src/torchmetrics/__init__.py

@@ -62,6 +62,7 @@
     HingeLoss,
     JaccardIndex,
     MatthewsCorrCoef,
+    NegativePredictiveValue,
     Precision,
     PrecisionAtFixedRecall,
     PrecisionRecallCurve,
@@ -207,6 +208,7 @@
     "MultiScaleStructuralSimilarityIndexMeasure",
     "MultioutputWrapper",
     "MultitaskWrapper",
+    "NegativePredictiveValue",
     "NormalizedRootMeanSquaredError",
     "PanopticQuality",
     "PeakSignalNoiseRatio",

src/torchmetrics/classification/__init__.py

@@ -63,6 +63,12 @@
     MulticlassMatthewsCorrCoef,
     MultilabelMatthewsCorrCoef,
 )
+from torchmetrics.classification.negative_predictive_value import (
+    BinaryNegativePredictiveValue,
+    MulticlassNegativePredictiveValue,
+    MultilabelNegativePredictiveValue,
+    NegativePredictiveValue,
+)
 from torchmetrics.classification.precision_fixed_recall import (
     BinaryPrecisionAtFixedRecall,
     MulticlassPrecisionAtFixedRecall,
@@ -217,4 +223,8 @@
     "MulticlassSensitivityAtSpecificity",
     "MultilabelSensitivityAtSpecificity",
     "SensitivityAtSpecificity",
+    "BinaryNegativePredictiveValue",
+    "MulticlassNegativePredictiveValue",
+    "MultilabelNegativePredictiveValue",
+    "NegativePredictiveValue",
 ]