[tests] further fix Tester object has no attribute '_testMethodName' (#35781)

* bug fix

* update with more cases

* more entries

* Fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

tests/models/albert/test_modeling_flax_albert.py

@@ -36,7 +36,7 @@ if is_flax_available():
     )
 
 
-class FlaxAlbertModelTester(unittest.TestCase):
+class FlaxAlbertModelTester:
     def __init__(
         self,
         parent,
@@ -80,7 +80,6 @@ class FlaxAlbertModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/aria/test_image_processing_aria.py

@@ -35,7 +35,7 @@ if is_torch_available():
     import torch
 
 
-class AriaImageProcessingTester(unittest.TestCase):
+class AriaImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -55,7 +55,6 @@ class AriaImageProcessingTester(unittest.TestCase):
         do_convert_rgb=True,
         resample=PILImageResampling.BICUBIC,
     ):
-        super().__init__()
         self.size = size if size is not None else {"longest_edge": max_resolution}
         self.parent = parent
         self.batch_size = batch_size

tests/models/beit/test_modeling_flax_beit.py

@@ -36,7 +36,7 @@ if is_vision_available():
     from transformers import BeitImageProcessor
 
 
-class FlaxBeitModelTester(unittest.TestCase):
+class FlaxBeitModelTester:
     def __init__(
         self,
         parent,
@@ -79,7 +79,6 @@ class FlaxBeitModelTester(unittest.TestCase):
         # in BeiT, the seq length equals the number of patches + 1 (we add 1 for the [CLS] token)
         num_patches = (image_size // patch_size) ** 2
         self.seq_length = num_patches + 1
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])

tests/models/bert/test_modeling_flax_bert.py

@@ -35,7 +35,7 @@ if is_flax_available():
     )
 
 
-class FlaxBertModelTester(unittest.TestCase):
+class FlaxBertModelTester:
     def __init__(
         self,
         parent,
@@ -79,7 +79,6 @@ class FlaxBertModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/big_bird/test_modeling_flax_big_bird.py

@@ -35,7 +35,7 @@ if is_flax_available():
     )
 
 
-class FlaxBigBirdModelTester(unittest.TestCase):
+class FlaxBigBirdModelTester:
     def __init__(
         self,
         parent,
@@ -90,7 +90,6 @@ class FlaxBigBirdModelTester(unittest.TestCase):
         self.use_bias = use_bias
         self.block_size = block_size
         self.num_random_blocks = num_random_blocks
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/blip/test_image_processing_blip.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import BlipImageProcessor
 
 
-class BlipImageProcessingTester(unittest.TestCase):
+class BlipImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -43,7 +43,6 @@ class BlipImageProcessingTester(unittest.TestCase):
         image_std=[0.26862954, 0.26130258, 0.27577711],
         do_convert_rgb=True,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 20, "width": 20}
         self.parent = parent
         self.batch_size = batch_size

tests/models/bridgetower/test_image_processing_bridgetower.py

@@ -31,7 +31,7 @@ if is_vision_available():
     from transformers import BridgeTowerImageProcessor
 
 
-class BridgeTowerImageProcessingTester(unittest.TestCase):
+class BridgeTowerImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class BridgeTowerImageProcessingTester(unittest.TestCase):
         max_resolution=400,
         num_channels=3,
     ):
-        super().__init__()
         self.parent = parent
         self.do_resize = do_resize
         self.size = size if size is not None else {"shortest_edge": 288}

tests/models/chameleon/test_image_processing_chameleon.py

@@ -32,7 +32,7 @@ if is_vision_available():
     from transformers import ChameleonImageProcessor
 
 
-class ChameleonImageProcessingTester(unittest.TestCase):
+class ChameleonImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class ChameleonImageProcessingTester(unittest.TestCase):
         image_std=[1.0, 1.0, 1.0],
         do_convert_rgb=True,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 18}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/chinese_clip/test_image_processing_chinese_clip.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import ChineseCLIPImageProcessor
 
 
-class ChineseCLIPImageProcessingTester(unittest.TestCase):
+class ChineseCLIPImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -44,7 +44,6 @@ class ChineseCLIPImageProcessingTester(unittest.TestCase):
         image_std=[0.26862954, 0.26130258, 0.27577711],
         do_convert_rgb=True,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 224, "width": 224}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/convnext/test_image_processing_convnext.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import ConvNextImageProcessor
 
 
-class ConvNextImageProcessingTester(unittest.TestCase):
+class ConvNextImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -42,7 +42,6 @@ class ConvNextImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 20}
         self.parent = parent
         self.batch_size = batch_size

tests/models/deformable_detr/test_image_processing_deformable_detr.py

@@ -35,7 +35,7 @@ if is_vision_available():
     from transformers import DeformableDetrImageProcessor, DeformableDetrImageProcessorFast
 
 
-class DeformableDetrImageProcessingTester(unittest.TestCase):
+class DeformableDetrImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -52,7 +52,6 @@ class DeformableDetrImageProcessingTester(unittest.TestCase):
         rescale_factor=1 / 255,
         do_pad=True,
     ):
-        super().__init__()
         # by setting size["longest_edge"] > max_resolution we're effectively not testing this :p
         size = size if size is not None else {"shortest_edge": 18, "longest_edge": 1333}
         self.parent = parent

tests/models/deit/test_image_processing_deit.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import DeiTImageProcessor
 
 
-class DeiTImageProcessingTester(unittest.TestCase):
+class DeiTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -43,7 +43,6 @@ class DeiTImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 20, "width": 20}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
 

tests/models/detr/test_image_processing_detr.py

@@ -37,7 +37,7 @@ if is_vision_available():
         from transformers import DetrImageProcessorFast
 
 
-class DetrImageProcessingTester(unittest.TestCase):
+class DetrImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -54,7 +54,6 @@ class DetrImageProcessingTester(unittest.TestCase):
         image_std=[0.5, 0.5, 0.5],
         do_pad=True,
     ):
-        super().__init__()
         # by setting size["longest_edge"] > max_resolution we're effectively not testing this :p
         size = size if size is not None else {"shortest_edge": 18, "longest_edge": 1333}
         self.parent = parent

tests/models/distilbert/test_modeling_flax_distilbert.py

@@ -35,7 +35,7 @@ if is_flax_available():
     )
 
 
-class FlaxDistilBertModelTester(unittest.TestCase):
+class FlaxDistilBertModelTester:
     def __init__(
         self,
         parent,
@@ -79,7 +79,6 @@ class FlaxDistilBertModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/donut/test_image_processing_donut.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import DonutImageProcessor
 
 
-class DonutImageProcessingTester(unittest.TestCase):
+class DonutImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -51,7 +51,6 @@ class DonutImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/dpt/test_image_processing_dpt.py

@@ -34,7 +34,7 @@ if is_vision_available():
     from transformers import DPTImageProcessor
 
 
-class DPTImageProcessingTester(unittest.TestCase):
+class DPTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class DPTImageProcessingTester(unittest.TestCase):
         image_std=[0.5, 0.5, 0.5],
         do_reduce_labels=False,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/efficientnet/test_image_processing_efficientnet.py

@@ -28,7 +28,7 @@ if is_vision_available():
     from transformers import EfficientNetImageProcessor
 
 
-class EfficientNetImageProcessorTester(unittest.TestCase):
+class EfficientNetImageProcessorTester:
     def __init__(
         self,
         parent,
@@ -43,7 +43,6 @@ class EfficientNetImageProcessorTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/electra/test_modeling_flax_electra.py

@@ -21,7 +21,7 @@ if is_flax_available():
     )
 
 
-class FlaxElectraModelTester(unittest.TestCase):
+class FlaxElectraModelTester:
     def __init__(
         self,
         parent,
@@ -67,7 +67,6 @@ class FlaxElectraModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/flava/test_image_processing_flava.py

@@ -42,7 +42,7 @@ else:
     FLAVA_IMAGE_MEAN = FLAVA_IMAGE_STD = FLAVA_CODEBOOK_MEAN = FLAVA_CODEBOOK_STD = None
 
 
-class FlavaImageProcessingTester(unittest.TestCase):
+class FlavaImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -76,7 +76,6 @@ class FlavaImageProcessingTester(unittest.TestCase):
         codebook_image_mean=FLAVA_CODEBOOK_MEAN,
         codebook_image_std=FLAVA_CODEBOOK_STD,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 224, "width": 224}
         crop_size = crop_size if crop_size is not None else {"height": 224, "width": 224}
         codebook_size = codebook_size if codebook_size is not None else {"height": 112, "width": 112}

tests/models/glpn/test_image_processing_glpn.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import GLPNImageProcessor
 
 
-class GLPNImageProcessingTester(unittest.TestCase):
+class GLPNImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -46,7 +46,6 @@ class GLPNImageProcessingTester(unittest.TestCase):
         size_divisor=32,
         do_rescale=True,
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/idefics3/test_image_processing_idefics3.py

@@ -35,7 +35,7 @@ if is_torch_available():
     import torch
 
 
-class Idefics3ImageProcessingTester(unittest.TestCase):
+class Idefics3ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -58,7 +58,6 @@ class Idefics3ImageProcessingTester(unittest.TestCase):
         do_image_splitting=True,
         resample=PILImageResampling.LANCZOS,
     ):
-        super().__init__()
         self.size = size if size is not None else {"longest_edge": max_resolution}
         self.parent = parent
         self.batch_size = batch_size

tests/models/imagegpt/test_image_processing_imagegpt.py

@@ -38,7 +38,7 @@ if is_vision_available():
     from transformers import ImageGPTImageProcessor
 
 
-class ImageGPTImageProcessingTester(unittest.TestCase):
+class ImageGPTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -51,7 +51,6 @@ class ImageGPTImageProcessingTester(unittest.TestCase):
         size=None,
         do_normalize=True,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/instructblipvideo/test_image_processing_instrictblipvideo.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import InstructBlipVideoImageProcessor
 
 
-class InstructBlipVideoProcessingTester(unittest.TestCase):
+class InstructBlipVideoProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class InstructBlipVideoProcessingTester(unittest.TestCase):
         do_convert_rgb=True,
         frames=4,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/layoutlmv2/test_image_processing_layoutlmv2.py

@@ -28,7 +28,7 @@ if is_pytesseract_available():
     from transformers import LayoutLMv2ImageProcessor
 
 
-class LayoutLMv2ImageProcessingTester(unittest.TestCase):
+class LayoutLMv2ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -41,7 +41,6 @@ class LayoutLMv2ImageProcessingTester(unittest.TestCase):
         size=None,
         apply_ocr=True,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/layoutlmv3/test_image_processing_layoutlmv3.py

@@ -28,7 +28,7 @@ if is_pytesseract_available():
     from transformers import LayoutLMv3ImageProcessor
 
 
-class LayoutLMv3ImageProcessingTester(unittest.TestCase):
+class LayoutLMv3ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -41,7 +41,6 @@ class LayoutLMv3ImageProcessingTester(unittest.TestCase):
         size=None,
         apply_ocr=True,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/levit/test_image_processing_levit.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import LevitImageProcessor
 
 
-class LevitImageProcessingTester(unittest.TestCase):
+class LevitImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -43,7 +43,6 @@ class LevitImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 18}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/llava/test_image_processing_llava.py

@@ -31,7 +31,7 @@ if is_vision_available():
     from transformers import LlavaImageProcessor
 
 
-class LlavaImageProcessingTester(unittest.TestCase):
+class LlavaImageProcessingTester:
     def __init__(
         self,
         parent,

tests/models/mbart/test_modeling_flax_mbart.py

@@ -74,7 +74,7 @@ def prepare_mbart_inputs_dict(
     }
 
 
-class FlaxMBartModelTester(unittest.TestCase):
+class FlaxMBartModelTester:
     def __init__(
         self,
         parent,
@@ -116,7 +116,6 @@ class FlaxMBartModelTester(unittest.TestCase):
         self.bos_token_id = bos_token_id
         self.decoder_start_token_id = decoder_start_token_id
         self.initializer_range = initializer_range
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = np.clip(ids_tensor([self.batch_size, self.seq_length - 1], self.vocab_size), 3, self.vocab_size)

tests/models/mllama/test_image_processing_mllama.py

@@ -34,7 +34,7 @@ if is_torch_available():
     import torch
 
 
-class MllamaImageProcessingTester(unittest.TestCase):
+class MllamaImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -55,7 +55,6 @@ class MllamaImageProcessingTester(unittest.TestCase):
         do_pad=True,
         max_image_tiles=4,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 224, "width": 224}
         self.parent = parent
         self.batch_size = batch_size

tests/models/mobilenet_v1/test_image_processing_mobilenet_v1.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import MobileNetV1ImageProcessor
 
 
-class MobileNetV1ImageProcessingTester(unittest.TestCase):
+class MobileNetV1ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -40,7 +40,6 @@ class MobileNetV1ImageProcessingTester(unittest.TestCase):
         do_center_crop=True,
         crop_size=None,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 20}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/mobilenet_v2/test_image_processing_mobilenet_v2.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import MobileNetV2ImageProcessor
 
 
-class MobileNetV2ImageProcessingTester(unittest.TestCase):
+class MobileNetV2ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -40,7 +40,6 @@ class MobileNetV2ImageProcessingTester(unittest.TestCase):
         do_center_crop=True,
         crop_size=None,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 20}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/mobilevit/test_image_processing_mobilevit.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import MobileViTImageProcessor
 
 
-class MobileViTImageProcessingTester(unittest.TestCase):
+class MobileViTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -48,7 +48,6 @@ class MobileViTImageProcessingTester(unittest.TestCase):
         crop_size=None,
         do_flip_channel_order=True,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 20}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/nougat/test_image_processing_nougat.py

@@ -34,7 +34,7 @@ if is_vision_available():
     from transformers import NougatImageProcessor
 
 
-class NougatImageProcessingTester(unittest.TestCase):
+class NougatImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -53,7 +53,6 @@ class NougatImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 20, "width": 20}
         self.parent = parent
         self.batch_size = batch_size

tests/models/owlv2/test_image_processing_owlv2.py

@@ -31,7 +31,7 @@ if is_torch_available():
     import torch
 
 
-class Owlv2ImageProcessingTester(unittest.TestCase):
+class Owlv2ImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -47,7 +47,6 @@ class Owlv2ImageProcessingTester(unittest.TestCase):
         image_std=[0.26862954, 0.26130258, 0.27577711],
         do_convert_rgb=True,
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/owlvit/test_image_processing_owlvit.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import OwlViTImageProcessor
 
 
-class OwlViTImageProcessingTester(unittest.TestCase):
+class OwlViTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -44,7 +44,6 @@ class OwlViTImageProcessingTester(unittest.TestCase):
         image_std=[0.26862954, 0.26130258, 0.27577711],
         do_convert_rgb=True,
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/poolformer/test_image_processing_poolformer.py

@@ -25,7 +25,7 @@ if is_vision_available():
     from transformers import PoolFormerImageProcessor
 
 
-class PoolFormerImageProcessingTester(unittest.TestCase):
+class PoolFormerImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -41,7 +41,6 @@ class PoolFormerImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 30}
         crop_size = crop_size if crop_size is not None else {"height": 30, "width": 30}
         self.parent = parent

tests/models/pvt/test_image_processing_pvt.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import PvtImageProcessor
 
 
-class PvtImageProcessingTester(unittest.TestCase):
+class PvtImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -41,7 +41,6 @@ class PvtImageProcessingTester(unittest.TestCase):
         image_mean=[0.485, 0.456, 0.406],
         image_std=[0.229, 0.224, 0.225],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/regnet/test_modeling_flax_regnet.py

@@ -36,7 +36,7 @@ if is_vision_available():
     from transformers import AutoImageProcessor
 
 
-class FlaxRegNetModelTester(unittest.TestCase):
+class FlaxRegNetModelTester:
     def __init__(
         self,
         parent,
@@ -65,7 +65,6 @@ class FlaxRegNetModelTester(unittest.TestCase):
         self.num_labels = num_labels
         self.scope = scope
         self.num_stages = len(hidden_sizes)
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])

tests/models/resnet/test_modeling_flax_resnet.py

@@ -35,7 +35,7 @@ if is_vision_available():
     from transformers import AutoImageProcessor
 
 
-class FlaxResNetModelTester(unittest.TestCase):
+class FlaxResNetModelTester:
     def __init__(
         self,
         parent,
@@ -64,7 +64,6 @@ class FlaxResNetModelTester(unittest.TestCase):
         self.num_labels = num_labels
         self.scope = scope
         self.num_stages = len(hidden_sizes)
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])

tests/models/roberta/test_modeling_flax_roberta.py

@@ -34,7 +34,7 @@ if is_flax_available():
     )
 
 
-class FlaxRobertaModelTester(unittest.TestCase):
+class FlaxRobertaModelTester:
     def __init__(
         self,
         parent,
@@ -78,7 +78,6 @@ class FlaxRobertaModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/roberta_prelayernorm/test_modeling_flax_roberta_prelayernorm.py

@@ -37,7 +37,7 @@ if is_flax_available():
 
 
 # Copied from tests.models.roberta.test_modeling_flax_roberta.FlaxRobertaModelTester with Roberta->RobertaPreLayerNorm
-class FlaxRobertaPreLayerNormModelTester(unittest.TestCase):
+class FlaxRobertaPreLayerNormModelTester:
     def __init__(
         self,
         parent,
@@ -81,7 +81,6 @@ class FlaxRobertaPreLayerNormModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/roformer/test_modeling_flax_roformer.py

@@ -35,7 +35,7 @@ if is_flax_available():
     )
 
 
-class FlaxRoFormerModelTester(unittest.TestCase):
+class FlaxRoFormerModelTester:
     def __init__(
         self,
         parent,
@@ -79,7 +79,6 @@ class FlaxRoFormerModelTester(unittest.TestCase):
         self.type_sequence_label_size = type_sequence_label_size
         self.initializer_range = initializer_range
         self.num_choices = num_choices
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/models/rt_detr/test_image_processing_rt_detr.py

@@ -31,7 +31,7 @@ if is_torch_available():
     import torch
 
 
-class RTDetrImageProcessingTester(unittest.TestCase):
+class RTDetrImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -45,7 +45,6 @@ class RTDetrImageProcessingTester(unittest.TestCase):
         do_pad=False,
         return_tensors="pt",
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/siglip/test_image_processing_siglip.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import SiglipImageProcessor
 
 
-class SiglipImageProcessingTester(unittest.TestCase):
+class SiglipImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -43,7 +43,6 @@ class SiglipImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/superglue/test_image_processing_superglue.py

@@ -42,7 +42,7 @@ def random_tensor(size):
     return torch.rand(size)
 
 
-class SuperGlueImageProcessingTester(unittest.TestCase):
+class SuperGlueImageProcessingTester:
     def __init__(
         self,
         parent,

tests/models/swin2sr/test_image_processing_swin2sr.py

@@ -34,7 +34,7 @@ if is_vision_available():
     from transformers.image_transforms import get_image_size
 
 
-class Swin2SRImageProcessingTester(unittest.TestCase):
+class Swin2SRImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -48,7 +48,6 @@ class Swin2SRImageProcessingTester(unittest.TestCase):
         do_pad=True,
         pad_size=8,
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/textnet/test_image_processing_textnet.py

@@ -26,7 +26,7 @@ if is_vision_available():
     from transformers import TextNetImageProcessor
 
 
-class TextNetImageProcessingTester(unittest.TestCase):
+class TextNetImageProcessingTester:
     def __init__(
         self,
         parent,

tests/models/tvp/test_image_processing_tvp.py

@@ -35,7 +35,7 @@ if is_vision_available():
     from transformers import TvpImageProcessor
 
 
-class TvpImageProcessingTester(unittest.TestCase):
+class TvpImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -58,7 +58,6 @@ class TvpImageProcessingTester(unittest.TestCase):
         num_channels=3,
         num_frames=2,
     ):
-        super().__init__()
         self.do_resize = do_resize
         self.size = size
         self.do_center_crop = do_center_crop

tests/models/video_llava/test_image_processing_video_llava.py

@@ -34,7 +34,7 @@ if is_vision_available():
     from transformers import VideoLlavaImageProcessor
 
 
-class VideoLlavaImageProcessingTester(unittest.TestCase):
+class VideoLlavaImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -52,7 +52,6 @@ class VideoLlavaImageProcessingTester(unittest.TestCase):
         image_std=OPENAI_CLIP_STD,
         do_convert_rgb=True,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 20}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
         self.parent = parent

tests/models/videomae/test_image_processing_videomae.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import VideoMAEImageProcessor
 
 
-class VideoMAEImageProcessingTester(unittest.TestCase):
+class VideoMAEImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class VideoMAEImageProcessingTester(unittest.TestCase):
         image_std=[0.5, 0.5, 0.5],
         crop_size=None,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 18}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
 

tests/models/vilt/test_image_processing_vilt.py

@@ -30,7 +30,7 @@ if is_vision_available():
     from transformers import ViltImageProcessor
 
 
-class ViltImageProcessingTester(unittest.TestCase):
+class ViltImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -46,7 +46,6 @@ class ViltImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 30}
         self.parent = parent
         self.batch_size = batch_size

tests/models/vit/test_image_processing_vit.py

@@ -29,7 +29,7 @@ if is_torchvision_available():
     from transformers import ViTImageProcessorFast
 
 
-class ViTImageProcessingTester(unittest.TestCase):
+class ViTImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -44,7 +44,6 @@ class ViTImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size

tests/models/vit/test_modeling_flax_vit.py

@@ -30,7 +30,7 @@ if is_flax_available():
     from transformers.models.vit.modeling_flax_vit import FlaxViTForImageClassification, FlaxViTModel
 
 
-class FlaxViTModelTester(unittest.TestCase):
+class FlaxViTModelTester:
     def __init__(
         self,
         parent,
@@ -72,7 +72,6 @@ class FlaxViTModelTester(unittest.TestCase):
         # in ViT, the seq length equals the number of patches + 1 (we add 1 for the [CLS] token)
         num_patches = (image_size // patch_size) ** 2
         self.seq_length = num_patches + 1
-        super().__init__()
 
     def prepare_config_and_inputs(self):
         pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size])

tests/models/vitmatte/test_image_processing_vitmatte.py

@@ -35,7 +35,7 @@ if is_vision_available():
     from transformers import VitMatteImageProcessor
 
 
-class VitMatteImageProcessingTester(unittest.TestCase):
+class VitMatteImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -52,7 +52,6 @@ class VitMatteImageProcessingTester(unittest.TestCase):
         image_mean=[0.5, 0.5, 0.5],
         image_std=[0.5, 0.5, 0.5],
     ):
-        super().__init__()
         self.parent = parent
         self.batch_size = batch_size
         self.num_channels = num_channels

tests/models/vitpose/test_image_processing_vitpose.py

@@ -34,7 +34,7 @@ if is_vision_available():
     from transformers import VitPoseImageProcessor
 
 
-class VitPoseImageProcessingTester(unittest.TestCase):
+class VitPoseImageProcessingTester:
     def __init__(
         self,
         parent,

tests/models/vivit/test_image_processing_vivit.py

@@ -33,7 +33,7 @@ if is_vision_available():
     from transformers import VivitImageProcessor
 
 
-class VivitImageProcessingTester(unittest.TestCase):
+class VivitImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -50,7 +50,6 @@ class VivitImageProcessingTester(unittest.TestCase):
         image_std=[0.5, 0.5, 0.5],
         crop_size=None,
     ):
-        super().__init__()
         size = size if size is not None else {"shortest_edge": 18}
         crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
 

tests/models/zoedepth/test_image_processing_zoedepth.py

@@ -28,7 +28,7 @@ if is_vision_available():
     from transformers import ZoeDepthImageProcessor
 
 
-class ZoeDepthImageProcessingTester(unittest.TestCase):
+class ZoeDepthImageProcessingTester:
     def __init__(
         self,
         parent,
@@ -46,7 +46,6 @@ class ZoeDepthImageProcessingTester(unittest.TestCase):
         image_std=[0.5, 0.5, 0.5],
         do_pad=False,
     ):
-        super().__init__()
         size = size if size is not None else {"height": 18, "width": 18}
         self.parent = parent
         self.batch_size = batch_size