Add class properties with warnings

src/transformers/models/beit/image_processing_beit.py

@@ -131,6 +131,15 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_STANDARD_STD
         self.do_reduce_labels = do_reduce_labels
 
+    @property
+    def reduce_labels(self) -> bool:
+        warnings.warn(
+            "The `reduce_labels` property is deprecated and will be removed in v4.27. Please use"
+            " `do_reduce_labels` instead.",
+            FutureWarning,
+        )
+        return self.do_reduce_labels
+
     @classmethod
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
         """

src/transformers/models/conditional_detr/image_processing_conditional_detr.py

@@ -815,6 +815,16 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_DEFAULT_STD
         self.do_pad = do_pad
 
+    @property
+    # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.max_size
+    def max_size(self):
+        warnings.warn(
+            "The `max_size` parameter is deprecated and will be removed in v4.27. "
+            "Please specify in `size['longest_edge'] instead`.",
+            FutureWarning,
+        )
+        return self.size["longest_edge"]
+
     @classmethod
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.from_dict with Detr->ConditionalDetr
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):

src/transformers/models/deformable_detr/image_processing_deformable_detr.py

@@ -813,6 +813,16 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_DEFAULT_STD
         self.do_pad = do_pad
 
+    @property
+    # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.max_size
+    def max_size(self):
+        warnings.warn(
+            "The `max_size` parameter is deprecated and will be removed in v4.27. "
+            "Please specify in `size['longest_edge'] instead`.",
+            FutureWarning,
+        )
+        return self.size["longest_edge"]
+
     @classmethod
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.from_dict with Detr->DeformableDetr
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):

src/transformers/models/detr/image_processing_detr.py

@@ -797,6 +797,15 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_DEFAULT_STD
         self.do_pad = do_pad
 
+    @property
+    def max_size(self):
+        warnings.warn(
+            "The `max_size` parameter is deprecated and will be removed in v4.27. "
+            "Please specify in `size['longest_edge'] instead`.",
+            FutureWarning,
+        )
+        return self.size["longest_edge"]
+
     @classmethod
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
         """

src/transformers/models/segformer/image_processing_segformer.py

@@ -119,6 +119,15 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_DEFAULT_STD
         self.do_reduce_labels = do_reduce_labels
 
+    @property
+    def reduce_labels(self):
+        warnings.warn(
+            "The `reduce_labels` property is deprecated and will be removed in a v4.27. Please use "
+            "`do_reduce_labels` instead.",
+            FutureWarning,
+        )
+        return self.do_reduce_labels
+
     @classmethod
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
         """

src/transformers/models/yolos/image_processing_yolos.py

@@ -725,6 +725,16 @@ def __init__(
         self.image_std = image_std if image_std is not None else IMAGENET_DEFAULT_STD
         self.do_pad = do_pad
 
+    @property
+    # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.max_size
+    def max_size(self):
+        warnings.warn(
+            "The `max_size` parameter is deprecated and will be removed in v4.27. "
+            "Please specify in `size['longest_edge'] instead`.",
+            FutureWarning,
+        )
+        return self.size["longest_edge"]
+
     @classmethod
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.from_dict with Detr->Yolos
     def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):

tests/models/beit/test_image_processing_beit.py

@@ -353,7 +353,7 @@ def test_reduce_labels(self):
         self.assertTrue(encoding["labels"].min().item() >= 0)
         self.assertTrue(encoding["labels"].max().item() <= 150)
 
-        image_processing.reduce_labels = True
+        image_processing.do_reduce_labels = True
         encoding = image_processing(image, map, return_tensors="pt")
         self.assertTrue(encoding["labels"].min().item() >= 0)
         self.assertTrue(encoding["labels"].max().item() <= 255)

tests/models/segformer/test_image_processing_segformer.py

@@ -339,7 +339,7 @@ def test_reduce_labels(self):
         self.assertTrue(encoding["labels"].min().item() >= 0)
         self.assertTrue(encoding["labels"].max().item() <= 150)
 
-        image_processing.reduce_labels = True
+        image_processing.do_reduce_labels = True
         encoding = image_processing(image, map, return_tensors="pt")
         self.assertTrue(encoding["labels"].min().item() >= 0)
         self.assertTrue(encoding["labels"].max().item() <= 255)