Update old existing feature extractor references (#24552)

* Update old existing feature extractor references

* Typo

* Apply suggestions from code review

* Apply suggestions from code review

* Apply suggestions from code review

* Address comments from review - update 'feature extractor'
Co-authored by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>

tests/models/maskformer/test_image_processing_maskformer.py

@@ -574,9 +574,9 @@ class MaskFormerImageProcessingTest(ImageProcessingSavingTestMixin, unittest.Tes
         self.assertEqual(segmentation[0].shape, target_sizes[0])
 
     def test_post_process_instance_segmentation(self):
-        feature_extractor = self.image_processing_class(num_labels=self.image_processor_tester.num_classes)
+        image_processor = self.image_processing_class(num_labels=self.image_processor_tester.num_classes)
         outputs = self.image_processor_tester.get_fake_maskformer_outputs()
-        segmentation = feature_extractor.post_process_instance_segmentation(outputs, threshold=0)
+        segmentation = image_processor.post_process_instance_segmentation(outputs, threshold=0)
 
         self.assertTrue(len(segmentation) == self.image_processor_tester.batch_size)
         for el in segmentation:
@@ -587,7 +587,7 @@ class MaskFormerImageProcessingTest(ImageProcessingSavingTestMixin, unittest.Tes
                 el["segmentation"].shape, (self.image_processor_tester.height, self.image_processor_tester.width)
             )
 
-        segmentation = feature_extractor.post_process_instance_segmentation(
+        segmentation = image_processor.post_process_instance_segmentation(
             outputs, threshold=0, return_binary_maps=True
         )
 

tests/models/maskformer/test_modeling_maskformer.py

@@ -35,7 +35,7 @@ if is_torch_available():
     from transformers import MaskFormerForInstanceSegmentation, MaskFormerModel
 
     if is_vision_available():
-        from transformers import MaskFormerFeatureExtractor
+        from transformers import MaskFormerImageProcessor
 
 if is_vision_available():
     from PIL import Image
@@ -326,18 +326,18 @@ def prepare_img():
 @slow
 class MaskFormerModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-small-coco")
+            MaskFormerImageProcessor.from_pretrained("facebook/maskformer-swin-small-coco")
             if is_vision_available()
             else None
         )
 
     def test_inference_no_head(self):
         model = MaskFormerModel.from_pretrained("facebook/maskformer-swin-small-coco").to(torch_device)
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").to(torch_device)
         inputs_shape = inputs["pixel_values"].shape
         # check size is divisible by 32
         self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
@@ -380,9 +380,9 @@ class MaskFormerModelIntegrationTest(unittest.TestCase):
             .to(torch_device)
             .eval()
         )
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").to(torch_device)
         inputs_shape = inputs["pixel_values"].shape
         # check size is divisible by 32
         self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
@@ -424,9 +424,9 @@ class MaskFormerModelIntegrationTest(unittest.TestCase):
             .to(torch_device)
             .eval()
         )
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").to(torch_device)
         inputs_shape = inputs["pixel_values"].shape
         # check size is divisible by 32
         self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
@@ -460,9 +460,9 @@ class MaskFormerModelIntegrationTest(unittest.TestCase):
             .to(torch_device)
             .eval()
         )
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
 
-        inputs = feature_extractor(
+        inputs = image_processor(
             [np.zeros((3, 800, 1333)), np.zeros((3, 800, 1333))],
             segmentation_maps=[np.zeros((384, 384)).astype(np.float32), np.zeros((384, 384)).astype(np.float32)],
             return_tensors="pt",

tests/models/mgp_str/test_processor_mgp_str.py

@@ -64,7 +64,7 @@ class MgpstrProcessorTest(unittest.TestCase):
         image_processor_map = {
             "do_normalize": False,
             "do_resize": True,
-            "feature_extractor_type": "ViTFeatureExtractor",
+            "image_processor_type": "ViTImageProcessor",
             "resample": 3,
             "size": {"height": 32, "width": 128},
         }

tests/models/mobilenet_v1/test_modeling_mobilenet_v1.py

@@ -37,7 +37,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import MobileNetV1FeatureExtractor
+    from transformers import MobileNetV1ImageProcessor
 
 
 class MobileNetV1ConfigTester(ConfigTester):
@@ -240,20 +240,18 @@ def prepare_img():
 @require_vision
 class MobileNetV1ModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            MobileNetV1FeatureExtractor.from_pretrained("google/mobilenet_v1_1.0_224")
-            if is_vision_available()
-            else None
+            MobileNetV1ImageProcessor.from_pretrained("google/mobilenet_v1_1.0_224") if is_vision_available() else None
         )
 
     @slow
     def test_inference_image_classification_head(self):
         model = MobileNetV1ForImageClassification.from_pretrained("google/mobilenet_v1_1.0_224").to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/mobilenet_v2/test_modeling_mobilenet_v2.py

@@ -37,7 +37,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import MobileNetV2FeatureExtractor
+    from transformers import MobileNetV2ImageProcessor
 
 
 class MobileNetV2ConfigTester(ConfigTester):
@@ -295,20 +295,18 @@ def prepare_img():
 @require_vision
 class MobileNetV2ModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            MobileNetV2FeatureExtractor.from_pretrained("google/mobilenet_v2_1.0_224")
-            if is_vision_available()
-            else None
+            MobileNetV2ImageProcessor.from_pretrained("google/mobilenet_v2_1.0_224") if is_vision_available() else None
         )
 
     @slow
     def test_inference_image_classification_head(self):
         model = MobileNetV2ForImageClassification.from_pretrained("google/mobilenet_v2_1.0_224").to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():
@@ -327,10 +325,10 @@ class MobileNetV2ModelIntegrationTest(unittest.TestCase):
         model = MobileNetV2ForSemanticSegmentation.from_pretrained("google/deeplabv3_mobilenet_v2_1.0_513")
         model = model.to(torch_device)
 
-        feature_extractor = MobileNetV2FeatureExtractor.from_pretrained("google/deeplabv3_mobilenet_v2_1.0_513")
+        image_processor = MobileNetV2ImageProcessor.from_pretrained("google/deeplabv3_mobilenet_v2_1.0_513")
 
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/mobilevit/test_modeling_mobilevit.py

@@ -37,7 +37,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import MobileViTFeatureExtractor
+    from transformers import MobileViTImageProcessor
 
 
 class MobileViTConfigTester(ConfigTester):
@@ -298,16 +298,16 @@ def prepare_img():
 @require_vision
 class MobileViTModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
-        return MobileViTFeatureExtractor.from_pretrained("apple/mobilevit-xx-small") if is_vision_available() else None
+    def default_image_processor(self):
+        return MobileViTImageProcessor.from_pretrained("apple/mobilevit-xx-small") if is_vision_available() else None
 
     @slow
     def test_inference_image_classification_head(self):
         model = MobileViTForImageClassification.from_pretrained("apple/mobilevit-xx-small").to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():
@@ -326,10 +326,10 @@ class MobileViTModelIntegrationTest(unittest.TestCase):
         model = MobileViTForSemanticSegmentation.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
         model = model.to(torch_device)
 
-        feature_extractor = MobileViTFeatureExtractor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
+        image_processor = MobileViTImageProcessor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
 
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():
@@ -356,10 +356,10 @@ class MobileViTModelIntegrationTest(unittest.TestCase):
         model = MobileViTForSemanticSegmentation.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
         model = model.to(torch_device)
 
-        feature_extractor = MobileViTFeatureExtractor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
+        image_processor = MobileViTImageProcessor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
 
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():
@@ -367,10 +367,10 @@ class MobileViTModelIntegrationTest(unittest.TestCase):
 
         outputs.logits = outputs.logits.detach().cpu()
 
-        segmentation = feature_extractor.post_process_semantic_segmentation(outputs=outputs, target_sizes=[(50, 60)])
+        segmentation = image_processor.post_process_semantic_segmentation(outputs=outputs, target_sizes=[(50, 60)])
         expected_shape = torch.Size((50, 60))
         self.assertEqual(segmentation[0].shape, expected_shape)
 
-        segmentation = feature_extractor.post_process_semantic_segmentation(outputs=outputs)
+        segmentation = image_processor.post_process_semantic_segmentation(outputs=outputs)
         expected_shape = torch.Size((32, 32))
         self.assertEqual(segmentation[0].shape, expected_shape)

tests/models/mobilevit/test_modeling_tf_mobilevit.py

@@ -40,7 +40,7 @@ if is_tf_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import MobileViTFeatureExtractor
+    from transformers import MobileViTImageProcessor
 
 
 class TFMobileViTConfigTester(ConfigTester):
@@ -381,9 +381,9 @@ class TFMobileViTModelIntegrationTest(unittest.TestCase):
     def test_inference_image_classification_head(self):
         model = TFMobileViTForImageClassification.from_pretrained("apple/mobilevit-xx-small")
 
-        feature_extractor = MobileViTFeatureExtractor.from_pretrained("apple/mobilevit-xx-small")
+        image_processor = MobileViTImageProcessor.from_pretrained("apple/mobilevit-xx-small")
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="tf")
+        inputs = image_processor(images=image, return_tensors="tf")
 
         # forward pass
         outputs = model(**inputs, training=False)
@@ -401,10 +401,10 @@ class TFMobileViTModelIntegrationTest(unittest.TestCase):
         # `from_pt` will be removed
         model = TFMobileViTForSemanticSegmentation.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
 
-        feature_extractor = MobileViTFeatureExtractor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
+        image_processor = MobileViTImageProcessor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
 
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="tf")
+        inputs = image_processor(images=image, return_tensors="tf")
 
         # forward pass
         outputs = model(inputs.pixel_values, training=False)

tests/models/nat/test_modeling_nat.py

@@ -364,16 +364,16 @@ class NatModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
 @require_torch
 class NatModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return AutoImageProcessor.from_pretrained("shi-labs/nat-mini-in1k-224") if is_vision_available() else None
 
     @slow
     def test_inference_image_classification_head(self):
         model = NatForImageClassification.from_pretrained("shi-labs/nat-mini-in1k-224").to(torch_device)
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
 
         image = Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png")
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/perceiver/test_modeling_perceiver.py

@@ -61,7 +61,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import PerceiverFeatureExtractor
+    from transformers import PerceiverImageProcessor
 
 
 class PerceiverModelTester:
@@ -899,13 +899,13 @@ class PerceiverModelIntegrationTest(unittest.TestCase):
 
     @slow
     def test_inference_image_classification(self):
-        feature_extractor = PerceiverFeatureExtractor()
+        image_processor = PerceiverImageProcessor()
         model = PerceiverForImageClassificationLearned.from_pretrained("deepmind/vision-perceiver-learned")
         model.to(torch_device)
 
         # prepare inputs
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").pixel_values.to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").pixel_values.to(torch_device)
         input_mask = None
 
         # forward pass
@@ -923,13 +923,13 @@ class PerceiverModelIntegrationTest(unittest.TestCase):
 
     @slow
     def test_inference_image_classification_fourier(self):
-        feature_extractor = PerceiverFeatureExtractor()
+        image_processor = PerceiverImageProcessor()
         model = PerceiverForImageClassificationFourier.from_pretrained("deepmind/vision-perceiver-fourier")
         model.to(torch_device)
 
         # prepare inputs
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").pixel_values.to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").pixel_values.to(torch_device)
         input_mask = None
 
         # forward pass
@@ -947,13 +947,13 @@ class PerceiverModelIntegrationTest(unittest.TestCase):
 
     @slow
     def test_inference_image_classification_conv(self):
-        feature_extractor = PerceiverFeatureExtractor()
+        image_processor = PerceiverImageProcessor()
         model = PerceiverForImageClassificationConvProcessing.from_pretrained("deepmind/vision-perceiver-conv")
         model.to(torch_device)
 
         # prepare inputs
         image = prepare_img()
-        inputs = feature_extractor(image, return_tensors="pt").pixel_values.to(torch_device)
+        inputs = image_processor(image, return_tensors="pt").pixel_values.to(torch_device)
         input_mask = None
 
         # forward pass

tests/models/poolformer/test_modeling_poolformer.py

@@ -37,7 +37,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import PoolFormerFeatureExtractor
+    from transformers import PoolFormerImageProcessor
 
 
 class PoolFormerConfigTester(ConfigTester):
@@ -237,10 +237,10 @@ def prepare_img():
 class PoolFormerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_classification_head(self):
-        feature_extractor = PoolFormerFeatureExtractor()
+        image_processor = PoolFormerImageProcessor()
         model = PoolFormerForImageClassification.from_pretrained("sail/poolformer_s12").to(torch_device)
 
-        inputs = feature_extractor(images=prepare_img(), return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=prepare_img(), return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/regnet/test_modeling_flax_regnet.py

@@ -33,7 +33,7 @@ if is_flax_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class FlaxRegNetModelTester(unittest.TestCase):
@@ -215,16 +215,16 @@ def prepare_img():
 @require_flax
 class FlaxRegNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
-        return AutoFeatureExtractor.from_pretrained("facebook/regnet-y-040") if is_vision_available() else None
+    def default_image_processor(self):
+        return AutoImageProcessor.from_pretrained("facebook/regnet-y-040") if is_vision_available() else None
 
     @slow
     def test_inference_image_classification_head(self):
         model = FlaxRegNetForImageClassification.from_pretrained("facebook/regnet-y-040")
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="np")
+        inputs = image_processor(images=image, return_tensors="np")
 
         outputs = model(**inputs)
 

tests/models/regnet/test_modeling_regnet.py

@@ -38,7 +38,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class RegNetModelTester:
@@ -248,9 +248,9 @@ def prepare_img():
 @require_vision
 class RegNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            AutoFeatureExtractor.from_pretrained(REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
+            AutoImageProcessor.from_pretrained(REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
             if is_vision_available()
             else None
         )
@@ -259,9 +259,9 @@ class RegNetModelIntegrationTest(unittest.TestCase):
     def test_inference_image_classification_head(self):
         model = RegNetForImageClassification.from_pretrained(REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0]).to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/regnet/test_modeling_tf_regnet.py

@@ -38,7 +38,7 @@ if is_tf_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class TFRegNetModelTester:
@@ -267,9 +267,9 @@ def prepare_img():
 @require_vision
 class RegNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            AutoFeatureExtractor.from_pretrained(TF_REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
+            AutoImageProcessor.from_pretrained(TF_REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
             if is_vision_available()
             else None
         )
@@ -278,9 +278,9 @@ class RegNetModelIntegrationTest(unittest.TestCase):
     def test_inference_image_classification_head(self):
         model = TFRegNetForImageClassification.from_pretrained(TF_REGNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="tf")
+        inputs = image_processor(images=image, return_tensors="tf")
 
         # forward pass
         outputs = model(**inputs, training=False)

tests/models/resnet/test_modeling_flax_resnet.py

@@ -32,7 +32,7 @@ if is_flax_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class FlaxResNetModelTester(unittest.TestCase):
@@ -206,16 +206,16 @@ def prepare_img():
 @require_flax
 class FlaxResNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
-        return AutoFeatureExtractor.from_pretrained("microsoft/resnet-50") if is_vision_available() else None
+    def default_image_processor(self):
+        return AutoImageProcessor.from_pretrained("microsoft/resnet-50") if is_vision_available() else None
 
     @slow
     def test_inference_image_classification_head(self):
         model = FlaxResNetForImageClassification.from_pretrained("microsoft/resnet-50")
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="np")
+        inputs = image_processor(images=image, return_tensors="np")
 
         outputs = model(**inputs)
 

tests/models/resnet/test_modeling_resnet.py

@@ -39,7 +39,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class ResNetModelTester:
@@ -301,9 +301,9 @@ def prepare_img():
 @require_vision
 class ResNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            AutoFeatureExtractor.from_pretrained(RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
+            AutoImageProcessor.from_pretrained(RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
             if is_vision_available()
             else None
         )
@@ -312,9 +312,9 @@ class ResNetModelIntegrationTest(unittest.TestCase):
     def test_inference_image_classification_head(self):
         model = ResNetForImageClassification.from_pretrained(RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0]).to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/resnet/test_modeling_tf_resnet.py

@@ -41,7 +41,7 @@ if is_tf_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class TFResNetModelTester:
@@ -229,9 +229,9 @@ def prepare_img():
 @require_vision
 class TFResNetModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            AutoFeatureExtractor.from_pretrained(TF_RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
+            AutoImageProcessor.from_pretrained(TF_RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
             if is_vision_available()
             else None
         )
@@ -240,9 +240,9 @@ class TFResNetModelIntegrationTest(unittest.TestCase):
     def test_inference_image_classification_head(self):
         model = TFResNetForImageClassification.from_pretrained(TF_RESNET_PRETRAINED_MODEL_ARCHIVE_LIST[0])
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="tf")
+        inputs = image_processor(images=image, return_tensors="tf")
 
         # forward pass
         outputs = model(**inputs)

tests/models/segformer/test_modeling_segformer.py

@@ -42,7 +42,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import SegformerFeatureExtractor
+    from transformers import SegformerImageProcessor
 
 
 class SegformerConfigTester(ConfigTester):
@@ -365,7 +365,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_segmentation_ade(self):
         # only resize + normalize
-        feature_extractor = SegformerFeatureExtractor(
+        image_processor = SegformerImageProcessor(
             image_scale=(512, 512), keep_ratio=False, align=False, do_random_crop=False
         )
         model = SegformerForSemanticSegmentation.from_pretrained("nvidia/segformer-b0-finetuned-ade-512-512").to(
@@ -373,7 +373,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
         )
 
         image = prepare_img()
-        encoded_inputs = feature_extractor(images=image, return_tensors="pt")
+        encoded_inputs = image_processor(images=image, return_tensors="pt")
         pixel_values = encoded_inputs.pixel_values.to(torch_device)
 
         with torch.no_grad():
@@ -394,7 +394,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_segmentation_city(self):
         # only resize + normalize
-        feature_extractor = SegformerFeatureExtractor(
+        image_processor = SegformerImageProcessor(
             image_scale=(512, 512), keep_ratio=False, align=False, do_random_crop=False
         )
         model = SegformerForSemanticSegmentation.from_pretrained(
@@ -402,7 +402,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
         ).to(torch_device)
 
         image = prepare_img()
-        encoded_inputs = feature_extractor(images=image, return_tensors="pt")
+        encoded_inputs = image_processor(images=image, return_tensors="pt")
         pixel_values = encoded_inputs.pixel_values.to(torch_device)
 
         with torch.no_grad():
@@ -423,7 +423,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_post_processing_semantic_segmentation(self):
         # only resize + normalize
-        feature_extractor = SegformerFeatureExtractor(
+        image_processor = SegformerImageProcessor(
             image_scale=(512, 512), keep_ratio=False, align=False, do_random_crop=False
         )
         model = SegformerForSemanticSegmentation.from_pretrained("nvidia/segformer-b0-finetuned-ade-512-512").to(
@@ -431,7 +431,7 @@ class SegformerModelIntegrationTest(unittest.TestCase):
         )
 
         image = prepare_img()
-        encoded_inputs = feature_extractor(images=image, return_tensors="pt")
+        encoded_inputs = image_processor(images=image, return_tensors="pt")
         pixel_values = encoded_inputs.pixel_values.to(torch_device)
 
         with torch.no_grad():
@@ -439,10 +439,10 @@ class SegformerModelIntegrationTest(unittest.TestCase):
 
         outputs.logits = outputs.logits.detach().cpu()
 
-        segmentation = feature_extractor.post_process_semantic_segmentation(outputs=outputs, target_sizes=[(500, 300)])
+        segmentation = image_processor.post_process_semantic_segmentation(outputs=outputs, target_sizes=[(500, 300)])
         expected_shape = torch.Size((500, 300))
         self.assertEqual(segmentation[0].shape, expected_shape)
 
-        segmentation = feature_extractor.post_process_semantic_segmentation(outputs=outputs)
+        segmentation = image_processor.post_process_semantic_segmentation(outputs=outputs)
         expected_shape = torch.Size((128, 128))
         self.assertEqual(segmentation[0].shape, expected_shape)

tests/models/segformer/test_modeling_tf_segformer.py

@@ -39,7 +39,7 @@ if is_tf_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import SegformerFeatureExtractor
+    from transformers import SegformerImageProcessor
 
 
 class TFSegformerConfigTester(ConfigTester):
@@ -454,13 +454,13 @@ class TFSegformerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_segmentation_ade(self):
         # only resize + normalize
-        feature_extractor = SegformerFeatureExtractor(
+        image_processor = SegformerImageProcessor(
             image_scale=(512, 512), keep_ratio=False, align=False, do_random_crop=False
         )
         model = TFSegformerForSemanticSegmentation.from_pretrained("nvidia/segformer-b0-finetuned-ade-512-512")
 
         image = prepare_img()
-        encoded_inputs = feature_extractor(images=image, return_tensors="tf")
+        encoded_inputs = image_processor(images=image, return_tensors="tf")
         pixel_values = encoded_inputs.pixel_values
 
         outputs = model(pixel_values, training=False)
@@ -480,7 +480,7 @@ class TFSegformerModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_segmentation_city(self):
         # only resize + normalize
-        feature_extractor = SegformerFeatureExtractor(
+        image_processor = SegformerImageProcessor(
             image_scale=(512, 512), keep_ratio=False, align=False, do_random_crop=False
         )
         model = TFSegformerForSemanticSegmentation.from_pretrained(
@@ -488,7 +488,7 @@ class TFSegformerModelIntegrationTest(unittest.TestCase):
         )
 
         image = prepare_img()
-        encoded_inputs = feature_extractor(images=image, return_tensors="tf")
+        encoded_inputs = image_processor(images=image, return_tensors="tf")
         pixel_values = encoded_inputs.pixel_values
 
         outputs = model(pixel_values, training=False)

tests/models/swiftformer/test_modeling_swiftformer.py

@@ -283,16 +283,16 @@ def prepare_img():
 @require_vision
 class SwiftFormerModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return ViTImageProcessor.from_pretrained("MBZUAI/swiftformer-xs") if is_vision_available() else None
 
     @slow
     def test_inference_image_classification_head(self):
         model = SwiftFormerForImageClassification.from_pretrained("MBZUAI/swiftformer-xs").to(torch_device)
 
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
         image = prepare_img()
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():

tests/models/swin/test_modeling_swin.py

@@ -39,7 +39,7 @@ if is_torch_available():
 if is_vision_available():
     from PIL import Image
 
-    from transformers import AutoFeatureExtractor
+    from transformers import AutoImageProcessor
 
 
 class SwinModelTester:
@@ -482,9 +482,9 @@ class SwinModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
 @require_torch
 class SwinModelIntegrationTest(unittest.TestCase):
     @cached_property
-    def default_feature_extractor(self):
+    def default_image_processor(self):
         return (
-            AutoFeatureExtractor.from_pretrained("microsoft/swin-tiny-patch4-window7-224")
+            AutoImageProcessor.from_pretrained("microsoft/swin-tiny-patch4-window7-224")
             if is_vision_available()
             else None
         )
@@ -492,10 +492,10 @@ class SwinModelIntegrationTest(unittest.TestCase):
     @slow
     def test_inference_image_classification_head(self):
         model = SwinForImageClassification.from_pretrained("microsoft/swin-tiny-patch4-window7-224").to(torch_device)
-        feature_extractor = self.default_feature_extractor
+        image_processor = self.default_image_processor
 
         image = Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png")
-        inputs = feature_extractor(images=image, return_tensors="pt").to(torch_device)
+        inputs = image_processor(images=image, return_tensors="pt").to(torch_device)
 
         # forward pass
         with torch.no_grad():