Change docs versionadded to 0.4.7

- additionally fixed docs for bleu and rouge

ignite/contrib/handlers/clearml_logger.py

@@ -303,7 +303,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/mlflow_logger.py

@@ -213,7 +213,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/neptune_logger.py

@@ -321,7 +321,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/polyaxon_logger.py

@@ -221,7 +221,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/tensorboard_logger.py

@@ -269,7 +269,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/tqdm_logger.py

@@ -100,7 +100,7 @@ class ProgressBar(BaseLogger):
         Due to `tqdm notebook bugs <https://github.com/tqdm/tqdm/issues/594>`_, bar format may be needed to be set
         to an empty string value.
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         `attach` now accepts an optional list of `state_attributes`
 
     """

ignite/contrib/handlers/visdom_logger.py

@@ -135,7 +135,7 @@ class VisdomLogger(BaseLogger):
                     output_transform=lambda loss: {"loss": loss}
                 )
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/contrib/handlers/wandb_logger.py

@@ -261,7 +261,7 @@ def global_step_transform(*args, **kwargs):
             def global_step_transform(engine, event_name):
                 return engine.state.get_event_attrib_value(event_name)
 
-    ..  versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         accepts an optional list of `state_attributes`
     """
 

ignite/distributed/auto.py

@@ -263,7 +263,7 @@ def auto_optim(optimizer: Optimizer, **kwargs: Any) -> Optimizer:
     .. versionchanged:: 0.4.2
         Added Horovod distributed optimizer.
 
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         Added kwargs to ``idist.auto_optim``.
     """
     bnd = idist.backend()

ignite/engine/__init__.py

@@ -84,7 +84,7 @@ def supervised_training_step(
             trainer = Engine(update_fn)
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         Added Gradient Accumulation.
     """
 
@@ -157,7 +157,7 @@ def supervised_training_step_amp(
             trainer = Engine(update_fn)
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         Added Gradient Accumulation.
     """
 
@@ -240,7 +240,7 @@ def supervised_training_step_apex(
             trainer = Engine(update_fn)
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         Added Gradient Accumulation.
     """
 
@@ -316,7 +316,7 @@ def supervised_training_step_tpu(
             trainer = Engine(update_fn)
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
        Added Gradient Accumulation argument for all supervised training methods.
     """
     try:
@@ -491,7 +491,7 @@ def output_transform_fn(x, y, y_pred, loss):
         - Added ``amp_mode`` argument for automatic mixed precision.
         - Added ``scaler`` argument for gradient scaling.
 
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
         Added Gradient Accumulation argument for all supervised training methods.
     """
 

ignite/handlers/checkpoint.py

@@ -259,7 +259,7 @@ class Checkpoint(Serializable):
         - Checkpoint can save model with same filename.
         - Added ``greater_or_equal`` argument.
 
-    .. versionchanged:: 0.5.0
+    .. versionchanged:: 0.4.7
 
         - `score_name` can be used to define `score_function` automatically without providing `score_function`.
         - `save_handler` automatically saves to disk if path to directory is provided.

ignite/handlers/param_scheduler.py

@@ -24,7 +24,7 @@ class BaseParamScheduler(metaclass=ABCMeta):
         save_history: whether to log the parameter values to
             `engine.state.param_history`, (default=False).
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 

ignite/handlers/state_param_scheduler.py

@@ -21,7 +21,7 @@ class StateParamScheduler(BaseParamScheduler):
         More precisely, whatever the state of the engine (newly created or used by another scheduler) the scheduler
         sets defined absolute values.
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 
@@ -178,7 +178,7 @@ def print_param():
             1.0
             0.9
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 
@@ -254,7 +254,7 @@ def print_param():
             0.64
             0.6
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
     """
 
     def __init__(
@@ -366,7 +366,7 @@ def print_param():
             0.7290...
             0.6561
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 
@@ -437,7 +437,7 @@ def print_param():
             0.6561
             0.5904...
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 
@@ -523,7 +523,7 @@ def print_param():
             0.7290...
             0.6561
 
-    .. versionadded:: 0.5.0
+    .. versionadded:: 0.4.7
 
     """
 

ignite/metrics/nlp/bleu.py

@@ -122,8 +122,10 @@ class Bleu(Metric):
             tensor(0.0393, dtype=torch.float64)
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
-        added ``average`` option to handle micro and macro averaging modes.
+    .. versionchanged:: 0.4.7
+
+        - ``update`` method has changed and now works on batch of inputs.
+        - added ``average`` option to handle micro and macro averaging modes.
     """
 
     def __init__(

ignite/metrics/nlp/rouge.py

@@ -368,8 +368,8 @@ class Rouge(Metric):
             {'Rouge-L-P': 0.6, 'Rouge-L-R': 0.5, 'Rouge-L-F': 0.5, 'Rouge-2-P': 0.5, 'Rouge-2-R': 0.4, 'Rouge-2-F': 0.4}
 
     .. versionadded:: 0.4.5
-    .. versionchanged:: 0.5.0
-        Changed input type to work on batch of inputs
+    .. versionchanged:: 0.4.7
+        ``update`` method has changed and now works on batch of inputs.
     """
 
     def __init__(