Deprecate low use models (#30781)

* Deprecate models
- graphormer
- time_series_transformer
- xlm_prophetnet
- qdqbert
- nat
- ernie_m
- tvlt
- nezha
- mega
- jukebox
- vit_hybrid
- x_clip
- deta
- speech_to_text_2
- efficientformer
- realm
- gptsan_japanese

* Fix up

* Fix speech2text2 imports

* Make sure message isn't indented

* Fix docstrings

* Correctly map for deprecated models from model_type

* Uncomment out

* Add back time series transformer and x-clip

* Import fix and fix-up

* Fix up with updated ruff

src/transformers/models/vit_hybrid/__init__.py → src/transformers/models/deprecated/vit_hybrid/__init__.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
+from ....utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
 
 
 _import_structure = {"configuration_vit_hybrid": ["ViTHybridConfig"]}

src/transformers/models/vit_hybrid/configuration_vit_hybrid.py → src/transformers/models/deprecated/vit_hybrid/configuration_vit_hybrid.py

@@ -14,10 +14,10 @@
 # limitations under the License.
 """ViT Hybrid model configuration"""
 
-from ...configuration_utils import PretrainedConfig
-from ...utils import logging
-from ..auto.configuration_auto import CONFIG_MAPPING
-from ..bit import BitConfig
+from ....configuration_utils import PretrainedConfig
+from ....utils import logging
+from ...auto.configuration_auto import CONFIG_MAPPING
+from ...bit import BitConfig
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/vit_hybrid/image_processing_vit_hybrid.py → src/transformers/models/deprecated/vit_hybrid/image_processing_vit_hybrid.py

@@ -18,14 +18,14 @@ from typing import Dict, List, Optional, Union
 
 import numpy as np
 
-from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
-from ...image_transforms import (
+from ....image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
+from ....image_transforms import (
     convert_to_rgb,
     get_resize_output_image_size,
     resize,
     to_channel_dimension_format,
 )
-from ...image_utils import (
+from ....image_utils import (
     OPENAI_CLIP_MEAN,
     OPENAI_CLIP_STD,
     ChannelDimension,
@@ -39,7 +39,7 @@ from ...image_utils import (
     validate_kwargs,
     validate_preprocess_arguments,
 )
-from ...utils import TensorType, is_vision_available, logging
+from ....utils import TensorType, is_vision_available, logging
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/vit_hybrid/modeling_vit_hybrid.py → src/transformers/models/deprecated/vit_hybrid/modeling_vit_hybrid.py

@@ -23,12 +23,12 @@ import torch.utils.checkpoint
 from torch import nn
 from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
-from ...activations import ACT2FN
-from ...modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling, ImageClassifierOutput
-from ...modeling_utils import PreTrainedModel
-from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
-from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
-from ...utils.backbone_utils import load_backbone
+from ....activations import ACT2FN
+from ....modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling, ImageClassifierOutput
+from ....modeling_utils import PreTrainedModel
+from ....pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
+from ....utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
+from ....utils.backbone_utils import load_backbone
 from .configuration_vit_hybrid import ViTHybridConfig
 
 

src/transformers/models/xlm_prophetnet/__init__.py → src/transformers/models/deprecated/xlm_prophetnet/__init__.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available, is_torch_available
+from ....utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available, is_torch_available
 
 
 _import_structure = {

src/transformers/models/xlm_prophetnet/configuration_xlm_prophetnet.py → src/transformers/models/deprecated/xlm_prophetnet/configuration_xlm_prophetnet.py

@@ -16,8 +16,8 @@
 
 from typing import Callable, Optional, Union
 
-from ...configuration_utils import PretrainedConfig
-from ...utils import logging
+from ....configuration_utils import PretrainedConfig
+from ....utils import logging
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/xlm_prophetnet/modeling_xlm_prophetnet.py → src/transformers/models/deprecated/xlm_prophetnet/modeling_xlm_prophetnet.py

@@ -25,10 +25,10 @@ import torch.utils.checkpoint
 from torch import Tensor, nn
 from torch.nn import LayerNorm
 
-from ...activations import ACT2FN
-from ...modeling_outputs import BaseModelOutput
-from ...modeling_utils import PreTrainedModel
-from ...utils import (
+from ....activations import ACT2FN
+from ....modeling_outputs import BaseModelOutput
+from ....modeling_utils import PreTrainedModel
+from ....utils import (
     ModelOutput,
     add_start_docstrings,
     add_start_docstrings_to_model_forward,

src/transformers/models/xlm_prophetnet/tokenization_xlm_prophetnet.py → src/transformers/models/deprecated/xlm_prophetnet/tokenization_xlm_prophetnet.py

@@ -18,8 +18,8 @@ import os
 from shutil import copyfile
 from typing import Any, Dict, List, Optional, Tuple
 
-from ...tokenization_utils import PreTrainedTokenizer
-from ...utils import logging
+from ....tokenization_utils import PreTrainedTokenizer
+from ....utils import logging
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/dinat/modeling_dinat.py

@@ -71,7 +71,6 @@ _IMAGE_CLASS_EXPECTED_OUTPUT = "tabby, tabby cat"
 
 
 @dataclass
-# Copied from transformers.models.nat.modeling_nat.NatEncoderOutput with Nat->Dinat
 class DinatEncoderOutput(ModelOutput):
     """
     Dinat encoder's outputs, with potential hidden states and attentions.
@@ -105,7 +104,6 @@ class DinatEncoderOutput(ModelOutput):
 
 
 @dataclass
-# Copied from transformers.models.nat.modeling_nat.NatModelOutput with Nat->Dinat
 class DinatModelOutput(ModelOutput):
     """
     Dinat model's outputs that also contains a pooling of the last hidden states.
@@ -142,7 +140,6 @@ class DinatModelOutput(ModelOutput):
 
 
 @dataclass
-# Copied from transformers.models.nat.modeling_nat.NatImageClassifierOutput with Nat->Dinat
 class DinatImageClassifierOutput(ModelOutput):
     """
     Dinat outputs for image classification.
@@ -178,7 +175,6 @@ class DinatImageClassifierOutput(ModelOutput):
     reshaped_hidden_states: Optional[Tuple[torch.FloatTensor, ...]] = None
 
 
-# Copied from transformers.models.nat.modeling_nat.NatEmbeddings with Nat->Dinat
 class DinatEmbeddings(nn.Module):
     """
     Construct the patch and position embeddings.
@@ -201,7 +197,6 @@ class DinatEmbeddings(nn.Module):
         return embeddings
 
 
-# Copied from transformers.models.nat.modeling_nat.NatPatchEmbeddings with Nat->Dinat
 class DinatPatchEmbeddings(nn.Module):
     """
     This class turns `pixel_values` of shape `(batch_size, num_channels, height, width)` into the initial
@@ -238,7 +233,6 @@ class DinatPatchEmbeddings(nn.Module):
         return embeddings
 
 
-# Copied from transformers.models.nat.modeling_nat.NatDownsampler with Nat->Dinat
 class DinatDownsampler(nn.Module):
     """
     Convolutional Downsampling Layer.
@@ -321,7 +315,6 @@ class NeighborhoodAttention(nn.Module):
 
         self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
 
-    # Copied from transformers.models.nat.modeling_nat.NeighborhoodAttention.transpose_for_scores with Nat->Dinat
     def transpose_for_scores(self, x):
         new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
         x = x.view(new_x_shape)
@@ -361,7 +354,6 @@ class NeighborhoodAttention(nn.Module):
         return outputs
 
 
-# Copied from transformers.models.nat.modeling_nat.NeighborhoodAttentionOutput
 class NeighborhoodAttentionOutput(nn.Module):
     def __init__(self, config, dim):
         super().__init__()
@@ -382,7 +374,6 @@ class NeighborhoodAttentionModule(nn.Module):
         self.output = NeighborhoodAttentionOutput(config, dim)
         self.pruned_heads = set()
 
-    # Copied from transformers.models.nat.modeling_nat.NeighborhoodAttentionModule.prune_heads
     def prune_heads(self, heads):
         if len(heads) == 0:
             return
@@ -401,7 +392,6 @@ class NeighborhoodAttentionModule(nn.Module):
         self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
         self.pruned_heads = self.pruned_heads.union(heads)
 
-    # Copied from transformers.models.nat.modeling_nat.NeighborhoodAttentionModule.forward
     def forward(
         self,
         hidden_states: torch.Tensor,
@@ -413,7 +403,6 @@ class NeighborhoodAttentionModule(nn.Module):
         return outputs
 
 
-# Copied from transformers.models.nat.modeling_nat.NatIntermediate with Nat->Dinat
 class DinatIntermediate(nn.Module):
     def __init__(self, config, dim):
         super().__init__()
@@ -429,7 +418,6 @@ class DinatIntermediate(nn.Module):
         return hidden_states
 
 
-# Copied from transformers.models.nat.modeling_nat.NatOutput with Nat->Dinat
 class DinatOutput(nn.Module):
     def __init__(self, config, dim):
         super().__init__()
@@ -539,7 +527,6 @@ class DinatStage(nn.Module):
 
         self.pointing = False
 
-    # Copied from transformers.models.nat.modeling_nat.NatStage.forward
     def forward(
         self,
         hidden_states: torch.Tensor,
@@ -582,7 +569,6 @@ class DinatEncoder(nn.Module):
             ]
         )
 
-    # Copied from transformers.models.nat.modeling_nat.NatEncoder.forward with Nat->Dinat
     def forward(
         self,
         hidden_states: torch.Tensor,
@@ -687,7 +673,6 @@ DINAT_INPUTS_DOCSTRING = r"""
     "The bare Dinat Model transformer outputting raw hidden-states without any specific head on top.",
     DINAT_START_DOCSTRING,
 )
-# Copied from transformers.models.nat.modeling_nat.NatModel with Nat->Dinat, NAT->DINAT
 class DinatModel(DinatPreTrainedModel):
     def __init__(self, config, add_pooling_layer=True):
         super().__init__(config)

src/transformers/utils/dummy_sentencepiece_objects.py

@@ -72,6 +72,13 @@ class ErnieMTokenizer(metaclass=DummyObject):
         requires_backends(self, ["sentencepiece"])
 
 
+class XLMProphetNetTokenizer(metaclass=DummyObject):
+    _backends = ["sentencepiece"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["sentencepiece"])
+
+
 class FNetTokenizer(metaclass=DummyObject):
     _backends = ["sentencepiece"]
 
@@ -233,13 +240,6 @@ class XGLMTokenizer(metaclass=DummyObject):
         requires_backends(self, ["sentencepiece"])
 
 
-class XLMProphetNetTokenizer(metaclass=DummyObject):
-    _backends = ["sentencepiece"]
-
-    def __init__(self, *args, **kwargs):
-        requires_backends(self, ["sentencepiece"])
-
-
 class XLMRobertaTokenizer(metaclass=DummyObject):
     _backends = ["sentencepiece"]
 

src/transformers/utils/dummy_tf_objects.py

@@ -1038,168 +1038,168 @@ class TFDeiTPreTrainedModel(metaclass=DummyObject):
         requires_backends(self, ["tf"])
 
 
-class TFAdaptiveEmbedding(metaclass=DummyObject):
+class TFEfficientFormerForImageClassification(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFTransfoXLForSequenceClassification(metaclass=DummyObject):
+class TFEfficientFormerForImageClassificationWithTeacher(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFTransfoXLLMHeadModel(metaclass=DummyObject):
+class TFEfficientFormerModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFTransfoXLMainLayer(metaclass=DummyObject):
+class TFEfficientFormerPreTrainedModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFTransfoXLModel(metaclass=DummyObject):
+class TFAdaptiveEmbedding(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFTransfoXLPreTrainedModel(metaclass=DummyObject):
+class TFTransfoXLForSequenceClassification(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertForMaskedLM(metaclass=DummyObject):
+class TFTransfoXLLMHeadModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertForMultipleChoice(metaclass=DummyObject):
+class TFTransfoXLMainLayer(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertForQuestionAnswering(metaclass=DummyObject):
+class TFTransfoXLModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertForSequenceClassification(metaclass=DummyObject):
+class TFTransfoXLPreTrainedModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertForTokenClassification(metaclass=DummyObject):
+class TFDistilBertForMaskedLM(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertMainLayer(metaclass=DummyObject):
+class TFDistilBertForMultipleChoice(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertModel(metaclass=DummyObject):
+class TFDistilBertForQuestionAnswering(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDistilBertPreTrainedModel(metaclass=DummyObject):
+class TFDistilBertForSequenceClassification(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRContextEncoder(metaclass=DummyObject):
+class TFDistilBertForTokenClassification(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRPretrainedContextEncoder(metaclass=DummyObject):
+class TFDistilBertMainLayer(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRPretrainedQuestionEncoder(metaclass=DummyObject):
+class TFDistilBertModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRPretrainedReader(metaclass=DummyObject):
+class TFDistilBertPreTrainedModel(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRQuestionEncoder(metaclass=DummyObject):
+class TFDPRContextEncoder(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFDPRReader(metaclass=DummyObject):
+class TFDPRPretrainedContextEncoder(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFEfficientFormerForImageClassification(metaclass=DummyObject):
+class TFDPRPretrainedQuestionEncoder(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFEfficientFormerForImageClassificationWithTeacher(metaclass=DummyObject):
+class TFDPRPretrainedReader(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFEfficientFormerModel(metaclass=DummyObject):
+class TFDPRQuestionEncoder(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tf"])
 
 
-class TFEfficientFormerPreTrainedModel(metaclass=DummyObject):
+class TFDPRReader(metaclass=DummyObject):
     _backends = ["tf"]
 
     def __init__(self, *args, **kwargs):

src/transformers/utils/dummy_tokenizers_objects.py

@@ -121,6 +121,13 @@ class DebertaV2TokenizerFast(metaclass=DummyObject):
         requires_backends(self, ["tokenizers"])
 
 
+class RealmTokenizerFast(metaclass=DummyObject):
+    _backends = ["tokenizers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["tokenizers"])
+
+
 class RetriBertTokenizerFast(metaclass=DummyObject):
     _backends = ["tokenizers"]
 
@@ -352,13 +359,6 @@ class Qwen2TokenizerFast(metaclass=DummyObject):
         requires_backends(self, ["tokenizers"])
 
 
-class RealmTokenizerFast(metaclass=DummyObject):
-    _backends = ["tokenizers"]
-
-    def __init__(self, *args, **kwargs):
-        requires_backends(self, ["tokenizers"])
-
-
 class ReformerTokenizerFast(metaclass=DummyObject):
     _backends = ["tokenizers"]
 

src/transformers/utils/dummy_vision_objects.py

@@ -142,49 +142,63 @@ class DetaImageProcessor(metaclass=DummyObject):
         requires_backends(self, ["vision"])
 
 
-class DetrFeatureExtractor(metaclass=DummyObject):
+class EfficientFormerImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class DetrImageProcessor(metaclass=DummyObject):
+class TvltImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class DonutFeatureExtractor(metaclass=DummyObject):
+class ViTHybridImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class DonutImageProcessor(metaclass=DummyObject):
+class DetrFeatureExtractor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class DPTFeatureExtractor(metaclass=DummyObject):
+class DetrImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class DPTImageProcessor(metaclass=DummyObject):
+class DonutFeatureExtractor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])
 
 
-class EfficientFormerImageProcessor(metaclass=DummyObject):
+class DonutImageProcessor(metaclass=DummyObject):
+    _backends = ["vision"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["vision"])
+
+
+class DPTFeatureExtractor(metaclass=DummyObject):
+    _backends = ["vision"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["vision"])
+
+
+class DPTImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
     def __init__(self, *args, **kwargs):
@@ -520,13 +534,6 @@ class Swin2SRImageProcessor(metaclass=DummyObject):
         requires_backends(self, ["vision"])
 
 
-class TvltImageProcessor(metaclass=DummyObject):
-    _backends = ["vision"]
-
-    def __init__(self, *args, **kwargs):
-        requires_backends(self, ["vision"])
-
-
 class TvpImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 
@@ -590,13 +597,6 @@ class ViTImageProcessor(metaclass=DummyObject):
         requires_backends(self, ["vision"])
 
 
-class ViTHybridImageProcessor(metaclass=DummyObject):
-    _backends = ["vision"]
-
-    def __init__(self, *args, **kwargs):
-        requires_backends(self, ["vision"])
-
-
 class VitMatteImageProcessor(metaclass=DummyObject):
     _backends = ["vision"]
 

tests/models/efficientformer/test_image_processing_efficientformer.py

-# coding=utf-8
-# Copyright 2021 HuggingFace Inc.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-
-import unittest
-
-from transformers.testing_utils import require_torch, require_vision
-from transformers.utils import is_vision_available
-
-from ...test_image_processing_common import ImageProcessingTestMixin, prepare_image_inputs
-
-
-if is_vision_available():
-    from transformers import ViTImageProcessor
-
-
-class EfficientFormerImageProcessorTester(unittest.TestCase):
-    def __init__(
-        self,
-        parent,
-        batch_size=13,
-        num_channels=3,
-        image_size=224,
-        min_resolution=30,
-        max_resolution=400,
-        do_resize=True,
-        size=None,
-        do_normalize=True,
-        image_mean=[0.5, 0.5, 0.5],
-        image_std=[0.5, 0.5, 0.5],
-    ):
-        size = size if size is not None else {"height": 18, "width": 18}
-        self.parent = parent
-        self.batch_size = batch_size
-        self.num_channels = num_channels
-        self.image_size = image_size
-        self.min_resolution = min_resolution
-        self.max_resolution = max_resolution
-        self.do_resize = do_resize
-        self.size = size
-        self.do_normalize = do_normalize
-        self.image_mean = image_mean
-        self.image_std = image_std
-
-    def prepare_image_processor_dict(self):
-        return {
-            "image_mean": self.image_mean,
-            "image_std": self.image_std,
-            "do_normalize": self.do_normalize,
-            "do_resize": self.do_resize,
-            "size": self.size,
-        }
-
-    def expected_output_image_shape(self, images):
-        return self.num_channels, self.size["height"], self.size["width"]
-
-    def prepare_image_inputs(self, equal_resolution=False, numpify=False, torchify=False):
-        return prepare_image_inputs(
-            batch_size=self.batch_size,
-            num_channels=self.num_channels,
-            min_resolution=self.min_resolution,
-            max_resolution=self.max_resolution,
-            equal_resolution=equal_resolution,
-            numpify=numpify,
-            torchify=torchify,
-        )
-
-
-@require_torch
-@require_vision
-class EfficientFormerImageProcessorTest(ImageProcessingTestMixin, unittest.TestCase):
-    image_processing_class = ViTImageProcessor if is_vision_available() else None
-
-    def setUp(self):
-        self.image_processor_tester = EfficientFormerImageProcessorTester(self)
-
-    @property
-    def image_processor_dict(self):
-        return self.image_processor_tester.prepare_image_processor_dict()
-
-    def test_image_proc_properties(self):
-        image_processor = self.image_processing_class(**self.image_processor_dict)
-        self.assertTrue(hasattr(image_processor, "image_mean"))
-        self.assertTrue(hasattr(image_processor, "image_std"))
-        self.assertTrue(hasattr(image_processor, "do_normalize"))
-        self.assertTrue(hasattr(image_processor, "do_resize"))
-        self.assertTrue(hasattr(image_processor, "size"))