Supporting `ImageProcessor` in place of `FeatureExtractor` for pipelines (#20851)

* Fixing the pipeline with image processor.

* Update the slow test.

* Using only the first image processor.

* Include exclusion mecanism for Image processor.

* Do not handle Gitconfig, deemed as a bug.

* Apply suggestions from code review
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* Remove `conversational` changes. They are not supposed to be here.

* Address first row of comments.

* Remove OneFormer modifications.
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

tests/pipelines/test_pipelines_text_generation.py

@@ -143,7 +143,7 @@ class TextGenerationPipelineTests(unittest.TestCase, metaclass=PipelineTestCaseM
             ],
         )
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         text_generator = TextGenerationPipeline(model=model, tokenizer=tokenizer)
         return text_generator, ["This is a test", "Another test"]
 

tests/pipelines/test_pipelines_token_classification.py

@@ -37,7 +37,7 @@ class TokenClassificationPipelineTests(unittest.TestCase, metaclass=PipelineTest
     model_mapping = MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING
     tf_model_mapping = TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         token_classifier = TokenClassificationPipeline(model=model, tokenizer=tokenizer)
         return token_classifier, ["A simple string", "A simple string that is quite a bit longer"]
 

tests/pipelines/test_pipelines_translation.py

@@ -34,7 +34,7 @@ class TranslationPipelineTests(unittest.TestCase, metaclass=PipelineTestCaseMeta
     model_mapping = MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
     tf_model_mapping = TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         if isinstance(model.config, MBartConfig):
             src_lang, tgt_lang = list(tokenizer.lang_code_to_id.keys())[:2]
             translator = TranslationPipeline(model=model, tokenizer=tokenizer, src_lang=src_lang, tgt_lang=tgt_lang)

tests/pipelines/test_pipelines_video_classification.py

@@ -35,7 +35,7 @@ from .test_pipelines_common import ANY, PipelineTestCaseMeta
 class VideoClassificationPipelineTests(unittest.TestCase, metaclass=PipelineTestCaseMeta):
     model_mapping = MODEL_FOR_VIDEO_CLASSIFICATION_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         example_video_filepath = hf_hub_download(
             repo_id="nateraw/video-demo", filename="archery.mp4", repo_type="dataset"
         )

tests/pipelines/test_pipelines_visual_question_answering.py

@@ -36,7 +36,7 @@ else:
 class VisualQuestionAnsweringPipelineTests(unittest.TestCase, metaclass=PipelineTestCaseMeta):
     model_mapping = MODEL_FOR_VISUAL_QUESTION_ANSWERING_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         vqa_pipeline = pipeline("visual-question-answering", model="hf-internal-testing/tiny-vilt-random-vqa")
         examples = [
             {

tests/pipelines/test_pipelines_zero_shot.py

@@ -30,7 +30,7 @@ class ZeroShotClassificationPipelineTests(unittest.TestCase, metaclass=PipelineT
     model_mapping = MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING
     tf_model_mapping = TF_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         classifier = ZeroShotClassificationPipeline(
             model=model, tokenizer=tokenizer, candidate_labels=["polics", "health"]
         )

tests/pipelines/test_pipelines_zero_shot_image_classification.py

@@ -37,7 +37,7 @@ class ZeroShotImageClassificationPipelineTests(unittest.TestCase, metaclass=Pipe
     # and only CLIP would be there for now.
     # model_mapping = {CLIPConfig: CLIPModel}
 
-    # def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    # def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
     #     if tokenizer is None:
     #         # Side effect of no Fast Tokenizer class for these model, so skipping
     #         # But the slow tokenizer test should still run as they're quite small

tests/pipelines/test_pipelines_zero_shot_object_detection.py

@@ -36,7 +36,7 @@ class ZeroShotObjectDetectionPipelineTests(unittest.TestCase, metaclass=Pipeline
 
     model_mapping = MODEL_FOR_ZERO_SHOT_OBJECT_DETECTION_MAPPING
 
-    def get_test_pipeline(self, model, tokenizer, feature_extractor):
+    def get_test_pipeline(self, model, tokenizer, feature_extractor, image_processor):
         object_detector = pipeline(
             "zero-shot-object-detection", model="hf-internal-testing/tiny-random-owlvit-object-detection"
         )