[Refactor] Classification 1/n

.github/actions/unittesting/action.yml

@@ -57,7 +57,7 @@ runs:
 
   - name: Unittests
     working-directory: ./tests
-    run: python -m pytest ${{ inputs.dirs }} --cov=torchmetrics --junitxml="$PYTEST_ARTEFACT.xml" --durations=50 ${{ inputs.test-timeout }}
+    run: python -m pytest -v --maxfail=5 ${{ inputs.dirs }} --cov=torchmetrics --junitxml="$PYTEST_ARTEFACT.xml" --durations=50 ${{ inputs.test-timeout }}
     shell: ${{ inputs.shell-type }}
 
   - name: Upload pytest results

CHANGELOG.md

@@ -19,7 +19,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Changed
 
--
+- Classification refactor (
+    [#1054](https://github.com/Lightning-AI/metrics/pull/1054),
+)
+
 
 -
 

docs/source/classification/confusion_matrix.rst

@@ -12,11 +12,56 @@ Confusion Matrix
 Module Interface
 ________________
 
+ConfusionMatrix
+^^^^^^^^^^^^^^^
+
 .. autoclass:: torchmetrics.ConfusionMatrix
     :noindex:
 
+BinaryConfusionMatrix
+^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.BinaryConfusionMatrix
+    :noindex:
+    :exclude-members: update, compute
+
+MulticlassConfusionMatrix
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.MulticlassConfusionMatrix
+    :noindex:
+    :exclude-members: update, compute
+
+MultilabelConfusionMatrix
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.MultilabelConfusionMatrix
+    :noindex:
+    :exclude-members: update, compute
+
 Functional Interface
 ____________________
 
+confusion_matrix
+^^^^^^^^^^^^^^^^
+
 .. autofunction:: torchmetrics.functional.confusion_matrix
     :noindex:
+
+binary_confusion_matrix
+^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.binary_confusion_matrix
+    :noindex:
+
+multiclass_confusion_matrix
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.multiclass_confusion_matrix
+    :noindex:
+
+multilabel_confusion_matrix
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.multilabel_confusion_matrix
+    :noindex:

docs/source/classification/stat_scores.rst

@@ -12,11 +12,56 @@ Stat Scores
 Module Interface
 ________________
 
+StatScores
+^^^^^^^^^^
+
 .. autoclass:: torchmetrics.StatScores
     :noindex:
 
+BinaryStatScores
+^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.BinaryStatScores
+    :noindex:
+    :exclude-members: update, compute
+
+MulticlassStatScores
+^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.MulticlassStatScores
+    :noindex:
+    :exclude-members: update, compute
+
+MultilabelStatScores
+^^^^^^^^^^^^^^^^^^^^
+
+.. autoclass:: torchmetrics.MultilabelStatScores
+    :noindex:
+    :exclude-members: update, compute
+
 Functional Interface
 ____________________
 
+stat_scores
+^^^^^^^^^^^
+
 .. autofunction:: torchmetrics.functional.stat_scores
     :noindex:
+
+binary_stat_scores
+^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.binary_stat_scores
+    :noindex:
+
+multiclass_stat_scores
+^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.multiclass_stat_scores
+    :noindex:
+
+multilabel_stat_scores
+^^^^^^^^^^^^^^^^^^^^^^
+
+.. autofunction:: torchmetrics.functional.multilabel_stat_scores
+    :noindex:

src/torchmetrics/__init__.py

@@ -26,6 +26,8 @@
     ROC,
     Accuracy,
     AveragePrecision,
+    BinaryConfusionMatrix,
+    BinaryStatScores,
     BinnedAveragePrecision,
     BinnedPrecisionRecallCurve,
     BinnedRecallAtFixedPrecision,
@@ -43,6 +45,10 @@
     LabelRankingAveragePrecision,
     LabelRankingLoss,
     MatthewsCorrCoef,
+    MulticlassConfusionMatrix,
+    MulticlassStatScores,
+    MultilabelConfusionMatrix,
+    MultilabelStatScores,
     Precision,
     PrecisionRecallCurve,
     Recall,
@@ -126,6 +132,9 @@
     "CHRFScore",
     "CohenKappa",
     "ConfusionMatrix",
+    "BinaryConfusionMatrix",
+    "MulticlassConfusionMatrix",
+    "MultilabelConfusionMatrix",
     "CosineSimilarity",
     "CoverageError",
     "Dice",
@@ -187,6 +196,9 @@
     "SQuAD",
     "StructuralSimilarityIndexMeasure",
     "StatScores",
+    "BinaryStatScores",
+    "MulticlassStatScores",
+    "MultilabelStatScores",
     "SumMetric",
     "SymmetricMeanAbsolutePercentageError",
     "TranslationEditRate",

src/torchmetrics/classification/__init__.py

@@ -20,7 +20,12 @@
 from torchmetrics.classification.binned_precision_recall import BinnedRecallAtFixedPrecision  # noqa: F401
 from torchmetrics.classification.calibration_error import CalibrationError  # noqa: F401
 from torchmetrics.classification.cohen_kappa import CohenKappa  # noqa: F401
-from torchmetrics.classification.confusion_matrix import ConfusionMatrix  # noqa: F401
+from torchmetrics.classification.confusion_matrix import (  # noqa: F401
+    BinaryConfusionMatrix,
+    ConfusionMatrix,
+    MulticlassConfusionMatrix,
+    MultilabelConfusionMatrix,
+)
 from torchmetrics.classification.dice import Dice  # noqa: F401
 from torchmetrics.classification.f_beta import F1Score, FBetaScore  # noqa: F401
 from torchmetrics.classification.hamming import HammingDistance  # noqa: F401
@@ -37,4 +42,9 @@
 )
 from torchmetrics.classification.roc import ROC  # noqa: F401
 from torchmetrics.classification.specificity import Specificity  # noqa: F401
-from torchmetrics.classification.stat_scores import StatScores  # noqa: F401
+from torchmetrics.classification.stat_scores import (  # noqa: F401
+    BinaryStatScores,
+    MulticlassStatScores,
+    MultilabelStatScores,
+    StatScores,
+)