Rework pipeline tests (#19366)

* Rework pipeline tests

* Try to fix Flax tests

* Try to put it before

* Use a new decorator instead

* Remove ignore marker since it doesn't work

* Filter pipeline tests

* Woopsie

* Use the fitlered list

* Clean up and fake modif

* Remove init

* Revert fake modif

tests/pipelines/test_pipelines_token_classification.py

@@ -25,14 +25,7 @@ from transformers import (
     pipeline,
 )
 from transformers.pipelines import AggregationStrategy, TokenClassificationArgumentHandler
-from transformers.testing_utils import (
-    is_pipeline_test,
-    nested_simplify,
-    require_tf,
-    require_torch,
-    require_torch_gpu,
-    slow,
-)
+from transformers.testing_utils import nested_simplify, require_tf, require_torch, require_torch_gpu, slow
 
 from .test_pipelines_common import ANY, PipelineTestCaseMeta
 
@@ -40,7 +33,6 @@ from .test_pipelines_common import ANY, PipelineTestCaseMeta
 VALID_INPUTS = ["A simple string", ["list of strings", "A simple string that is quite a bit longer"]]
 
 
-@is_pipeline_test
 class TokenClassificationPipelineTests(unittest.TestCase, metaclass=PipelineTestCaseMeta):
     model_mapping = MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING
     tf_model_mapping = TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING
@@ -770,7 +762,6 @@ class TokenClassificationPipelineTests(unittest.TestCase, metaclass=PipelineTest
         )
 
 
-@is_pipeline_test
 class TokenClassificationArgumentHandlerTestCase(unittest.TestCase):
     def setUp(self):
         self.args_parser = TokenClassificationArgumentHandler()