Updated docs versionadded/versionchanged to 0.5.4

Author: vfdev-5

ignite/base/mixins.py

@@ -13,7 +13,7 @@ class ResettableHandler(metaclass=ABCMeta):
     Subclasses must implement the :meth:`reset` method to clear any accumulated
     state, typically at the beginning of a training run.
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
     """
 
     @abstractmethod

ignite/handlers/early_stopping.py

@@ -52,7 +52,7 @@ def score_function(engine):
             # Note: the handler is attached to an *Evaluator* (runs one epoch on validation dataset).
             evaluator.add_event_handler(Events.COMPLETED, handler)
 
-    .. versionchanged:: 0.6.0
+    .. versionchanged:: 0.5.4
         Added `mode` parameter to support minimization in addition to maximization.
         Added `min_delta_mode` parameter to support both absolute and relative improvements.
 
@@ -132,7 +132,7 @@ def __call__(self, engine: Engine) -> None:
     def reset(self) -> None:
         """Reset the early stopping state, including the counter and best score.
 
-        .. versionadded:: 0.6.0
+        .. versionadded:: 0.5.4
         """
         self.counter = 0
         self.best_score = None
@@ -161,7 +161,7 @@ def attach(  # type: ignore[override]
             reset_event: The event on ``reset_engine`` that triggers the handler state reset.
                 Default is :attr:`~ignite.engine.events.Events.STARTED`.
 
-        .. versionadded:: 0.6.0
+        .. versionadded:: 0.5.4
         """
         engine.add_event_handler(event, self)
 

ignite/metrics/fairness/accuracy_difference.py

@@ -65,7 +65,7 @@ class SubgroupAccuracyDifference(SubgroupDifference):
 
             1.0
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
 
     References:
         - Verma & Rubin, `Fairness Definitions Explained

ignite/metrics/fairness/base.py

@@ -87,7 +87,7 @@ class SubgroupMetric(_SubgroupBase):
             form expected by the metric. Default is ``(y_pred, y, group_labels)``.
         device: specifies which device updates are accumulated on.
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
     """
 
     def compute(self) -> dict[Any, Any]:
@@ -110,7 +110,7 @@ class SubgroupDifference(_SubgroupBase):
         output_transform: a callable that is used to transform the engine output.
         device: specifies the computation device.
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
     """
 
     def compute(self) -> float:

ignite/metrics/fairness/demographic_parity.py

@@ -40,7 +40,7 @@ class SelectionRate(_BaseClassification):
 
             tensor([0.5000, 0.5000])
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
     """
 
     _state_dict_all_req_keys = ("_num_positives", "_num_examples")
@@ -164,7 +164,7 @@ class DemographicParityDifference(SubgroupDifference):
 
             1.0
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
 
     References:
         - Verma & Rubin, `Fairness Definitions Explained

ignite/metrics/matthews_corrcoef.py

@@ -72,7 +72,7 @@ class MatthewsCorrCoef(EpochMetric):
 
                 -0.33...
 
-        .. versionadded:: 0.6.0
+        .. versionadded:: 0.5.4
 
     """
 

ignite/metrics/rec_sys/hitrate.py

@@ -97,7 +97,7 @@ class HitRate(Metric):
 
             [0.0, 0.5, 0.5, 0.5]
 
-    .. versionadded:: 0.6.0
+    .. versionadded:: 0.5.4
     """
 
     required_output_keys = ("y_pred", "y")