Deprecate models (#24787)

* Deprecate some models

* Fix imports

* Fix inits too

* Remove tests

* Add deprecated banner to documentation

* Remove from init

* Fix auto classes

* Style

* Remote upgrade strategy 1

* Remove site package cache

* Revert this part

* Fix typo...

* Update utils

* Update docs/source/en/model_doc/bort.md
Co-authored-by: Lysandre Debut <lysandre.debut@reseau.eseo.fr>

* Address review comments

* With all files saved

---------
Co-authored-by: Lysandre Debut <lysandre.debut@reseau.eseo.fr>

docs/source/en/model_doc/bort.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # BORT
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The BORT model was proposed in [Optimal Subarchitecture Extraction for BERT](https://arxiv.org/abs/2010.10499) by

docs/source/en/model_doc/mctct.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # M-CTC-T
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The M-CTC-T model was proposed in [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) by Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert. The model is a 1B-param transformer encoder, with a CTC head over 8065 character labels and a language identification head over 60 language ID labels. It is trained on Common Voice (version 6.1, December 2020 release) and VoxPopuli. After training on Common Voice and VoxPopuli, the model is trained on Common Voice only. The labels are unnormalized character-level transcripts (punctuation and capitalization are not removed). The model takes as input Mel filterbank features from a 16Khz audio signal.

docs/source/en/model_doc/retribert.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # RetriBERT
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The RetriBERT model was proposed in the blog post [Explain Anything Like I'm Five: A Model for Open Domain Long Form

docs/source/en/model_doc/tapex.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # TAPEX
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The TAPEX model was proposed in [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu,

docs/source/en/model_doc/trajectory_transformer.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # Trajectory Transformer
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The Trajectory Transformer model was proposed in [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039)  by Michael Janner, Qiyang Li, Sergey Levine.

docs/source/en/model_doc/van.md

@@ -16,6 +16,15 @@ rendered properly in your Markdown viewer.
 
 # VAN
 
+<Tip warning={true}>
+
+This model is in maintenance mode only, so we won't accept any new PRs changing its code.
+
+If you run into any issues running this model, please reinstall the last version that supported this model: v4.30.0.
+You can do so by running the following command: `pip install -U transformers==4.30.0`.
+
+</Tip>
+
 ## Overview
 
 The VAN model was proposed in [Visual Attention Network](https://arxiv.org/abs/2202.09741) by Meng-Hao Guo, Cheng-Ze Lu, Zheng-Ning Liu, Ming-Ming Cheng, Shi-Min Hu.

src/transformers/__init__.py

@@ -202,7 +202,6 @@ _import_structure = {
         "Blip2VisionConfig",
     ],
     "models.bloom": ["BLOOM_PRETRAINED_CONFIG_ARCHIVE_MAP", "BloomConfig"],
-    "models.bort": [],
     "models.bridgetower": [
         "BRIDGETOWER_PRETRAINED_CONFIG_ARCHIVE_MAP",
         "BridgeTowerConfig",
@@ -263,6 +262,26 @@ _import_structure = {
     "models.decision_transformer": ["DECISION_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "DecisionTransformerConfig"],
     "models.deformable_detr": ["DEFORMABLE_DETR_PRETRAINED_CONFIG_ARCHIVE_MAP", "DeformableDetrConfig"],
     "models.deit": ["DEIT_PRETRAINED_CONFIG_ARCHIVE_MAP", "DeiTConfig"],
+    "models.deprecated": [],
+    "models.deprecated.bort": [],
+    "models.deprecated.mctct": [
+        "MCTCT_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "MCTCTConfig",
+        "MCTCTFeatureExtractor",
+        "MCTCTProcessor",
+    ],
+    "models.deprecated.mmbt": ["MMBTConfig"],
+    "models.deprecated.retribert": [
+        "RETRIBERT_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "RetriBertConfig",
+        "RetriBertTokenizer",
+    ],
+    "models.deprecated.tapex": ["TapexTokenizer"],
+    "models.deprecated.trajectory_transformer": [
+        "TRAJECTORY_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "TrajectoryTransformerConfig",
+    ],
+    "models.deprecated.van": ["VAN_PRETRAINED_CONFIG_ARCHIVE_MAP", "VanConfig"],
     "models.deta": ["DETA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DetaConfig"],
     "models.detr": ["DETR_PRETRAINED_CONFIG_ARCHIVE_MAP", "DetrConfig"],
     "models.dialogpt": [],
@@ -390,13 +409,11 @@ _import_structure = {
     "models.maskformer": ["MASKFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "MaskFormerConfig", "MaskFormerSwinConfig"],
     "models.mbart": ["MBartConfig"],
     "models.mbart50": [],
-    "models.mctct": ["MCTCT_PRETRAINED_CONFIG_ARCHIVE_MAP", "MCTCTConfig", "MCTCTFeatureExtractor", "MCTCTProcessor"],
     "models.mega": ["MEGA_PRETRAINED_CONFIG_ARCHIVE_MAP", "MegaConfig"],
     "models.megatron_bert": ["MEGATRON_BERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "MegatronBertConfig"],
     "models.megatron_gpt2": [],
     "models.mgp_str": ["MGP_STR_PRETRAINED_CONFIG_ARCHIVE_MAP", "MgpstrConfig", "MgpstrProcessor", "MgpstrTokenizer"],
     "models.mluke": [],
-    "models.mmbt": ["MMBTConfig"],
     "models.mobilebert": ["MOBILEBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "MobileBertConfig", "MobileBertTokenizer"],
     "models.mobilenet_v1": ["MOBILENET_V1_PRETRAINED_CONFIG_ARCHIVE_MAP", "MobileNetV1Config"],
     "models.mobilenet_v2": ["MOBILENET_V2_PRETRAINED_CONFIG_ARCHIVE_MAP", "MobileNetV2Config"],
@@ -451,7 +468,6 @@ _import_structure = {
     "models.regnet": ["REGNET_PRETRAINED_CONFIG_ARCHIVE_MAP", "RegNetConfig"],
     "models.rembert": ["REMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "RemBertConfig"],
     "models.resnet": ["RESNET_PRETRAINED_CONFIG_ARCHIVE_MAP", "ResNetConfig"],
-    "models.retribert": ["RETRIBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "RetriBertConfig", "RetriBertTokenizer"],
     "models.roberta": ["ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "RobertaConfig", "RobertaTokenizer"],
     "models.roberta_prelayernorm": ["ROBERTA_PRELAYERNORM_PRETRAINED_CONFIG_ARCHIVE_MAP", "RobertaPreLayerNormConfig"],
     "models.roc_bert": ["ROC_BERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoCBertConfig", "RoCBertTokenizer"],
@@ -498,17 +514,12 @@ _import_structure = {
     "models.t5": ["T5_PRETRAINED_CONFIG_ARCHIVE_MAP", "T5Config"],
     "models.table_transformer": ["TABLE_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "TableTransformerConfig"],
     "models.tapas": ["TAPAS_PRETRAINED_CONFIG_ARCHIVE_MAP", "TapasConfig", "TapasTokenizer"],
-    "models.tapex": ["TapexTokenizer"],
     "models.time_series_transformer": [
         "TIME_SERIES_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP",
         "TimeSeriesTransformerConfig",
     ],
     "models.timesformer": ["TIMESFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP", "TimesformerConfig"],
     "models.timm_backbone": ["TimmBackboneConfig"],
-    "models.trajectory_transformer": [
-        "TRAJECTORY_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP",
-        "TrajectoryTransformerConfig",
-    ],
     "models.transfo_xl": [
         "TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP",
         "TransfoXLConfig",
@@ -536,7 +547,6 @@ _import_structure = {
         "UniSpeechSatConfig",
     ],
     "models.upernet": ["UperNetConfig"],
-    "models.van": ["VAN_PRETRAINED_CONFIG_ARCHIVE_MAP", "VanConfig"],
     "models.videomae": ["VIDEOMAE_PRETRAINED_CONFIG_ARCHIVE_MAP", "VideoMAEConfig"],
     "models.vilt": [
         "VILT_PRETRAINED_CONFIG_ARCHIVE_MAP",
@@ -783,6 +793,7 @@ else:
     _import_structure["models.cpm"].append("CpmTokenizerFast")
     _import_structure["models.deberta"].append("DebertaTokenizerFast")
     _import_structure["models.deberta_v2"].append("DebertaV2TokenizerFast")
+    _import_structure["models.deprecated.retribert"].append("RetriBertTokenizerFast")
     _import_structure["models.distilbert"].append("DistilBertTokenizerFast")
     _import_structure["models.dpr"].extend(
         ["DPRContextEncoderTokenizerFast", "DPRQuestionEncoderTokenizerFast", "DPRReaderTokenizerFast"]
@@ -815,7 +826,6 @@ else:
     _import_structure["models.realm"].append("RealmTokenizerFast")
     _import_structure["models.reformer"].append("ReformerTokenizerFast")
     _import_structure["models.rembert"].append("RemBertTokenizerFast")
-    _import_structure["models.retribert"].append("RetriBertTokenizerFast")
     _import_structure["models.roberta"].append("RobertaTokenizerFast")
     _import_structure["models.roformer"].append("RoFormerTokenizerFast")
     _import_structure["models.splinter"].append("SplinterTokenizerFast")
@@ -1497,6 +1507,33 @@ else:
             "DeiTPreTrainedModel",
         ]
     )
+    _import_structure["models.deprecated.mctct"].extend(
+        [
+            "MCTCT_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "MCTCTForCTC",
+            "MCTCTModel",
+            "MCTCTPreTrainedModel",
+        ]
+    )
+    _import_structure["models.deprecated.mmbt"].extend(["MMBTForClassification", "MMBTModel", "ModalEmbeddings"])
+    _import_structure["models.deprecated.retribert"].extend(
+        ["RETRIBERT_PRETRAINED_MODEL_ARCHIVE_LIST", "RetriBertModel", "RetriBertPreTrainedModel"]
+    )
+    _import_structure["models.deprecated.trajectory_transformer"].extend(
+        [
+            "TRAJECTORY_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "TrajectoryTransformerModel",
+            "TrajectoryTransformerPreTrainedModel",
+        ]
+    )
+    _import_structure["models.deprecated.van"].extend(
+        [
+            "VAN_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "VanForImageClassification",
+            "VanModel",
+            "VanPreTrainedModel",
+        ]
+    )
     _import_structure["models.deta"].extend(
         [
             "DETA_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -2043,14 +2080,6 @@ else:
             "MBartPreTrainedModel",
         ]
     )
-    _import_structure["models.mctct"].extend(
-        [
-            "MCTCT_PRETRAINED_MODEL_ARCHIVE_LIST",
-            "MCTCTForCTC",
-            "MCTCTModel",
-            "MCTCTPreTrainedModel",
-        ]
-    )
     _import_structure["models.mega"].extend(
         [
             "MEGA_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -2087,7 +2116,6 @@ else:
             "MgpstrPreTrainedModel",
         ]
     )
-    _import_structure["models.mmbt"].extend(["MMBTForClassification", "MMBTModel", "ModalEmbeddings"])
     _import_structure["models.mobilebert"].extend(
         [
             "MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -2419,9 +2447,6 @@ else:
             "ResNetPreTrainedModel",
         ]
     )
-    _import_structure["models.retribert"].extend(
-        ["RETRIBERT_PRETRAINED_MODEL_ARCHIVE_LIST", "RetriBertModel", "RetriBertPreTrainedModel"]
-    )
     _import_structure["models.roberta"].extend(
         [
             "ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -2660,13 +2685,6 @@ else:
         ]
     )
     _import_structure["models.timm_backbone"].extend(["TimmBackbone"])
-    _import_structure["models.trajectory_transformer"].extend(
-        [
-            "TRAJECTORY_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
-            "TrajectoryTransformerModel",
-            "TrajectoryTransformerPreTrainedModel",
-        ]
-    )
     _import_structure["models.transfo_xl"].extend(
         [
             "TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -2727,14 +2745,6 @@ else:
             "UperNetPreTrainedModel",
         ]
     )
-    _import_structure["models.van"].extend(
-        [
-            "VAN_PRETRAINED_MODEL_ARCHIVE_LIST",
-            "VanForImageClassification",
-            "VanModel",
-            "VanPreTrainedModel",
-        ]
-    )
     _import_structure["models.videomae"].extend(
         [
             "VIDEOMAE_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -4187,6 +4197,24 @@ if TYPE_CHECKING:
     )
     from .models.deformable_detr import DEFORMABLE_DETR_PRETRAINED_CONFIG_ARCHIVE_MAP, DeformableDetrConfig
     from .models.deit import DEIT_PRETRAINED_CONFIG_ARCHIVE_MAP, DeiTConfig
+    from .models.deprecated.mctct import (
+        MCTCT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        MCTCTConfig,
+        MCTCTFeatureExtractor,
+        MCTCTProcessor,
+    )
+    from .models.deprecated.mmbt import MMBTConfig
+    from .models.deprecated.retribert import (
+        RETRIBERT_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        RetriBertConfig,
+        RetriBertTokenizer,
+    )
+    from .models.deprecated.tapex import TapexTokenizer
+    from .models.deprecated.trajectory_transformer import (
+        TRAJECTORY_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        TrajectoryTransformerConfig,
+    )
+    from .models.deprecated.van import VAN_PRETRAINED_CONFIG_ARCHIVE_MAP, VanConfig
     from .models.deta import DETA_PRETRAINED_CONFIG_ARCHIVE_MAP, DetaConfig
     from .models.detr import DETR_PRETRAINED_CONFIG_ARCHIVE_MAP, DetrConfig
     from .models.dinat import DINAT_PRETRAINED_CONFIG_ARCHIVE_MAP, DinatConfig
@@ -4304,11 +4332,9 @@ if TYPE_CHECKING:
     from .models.mask2former import MASK2FORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, Mask2FormerConfig
     from .models.maskformer import MASKFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, MaskFormerConfig, MaskFormerSwinConfig
     from .models.mbart import MBartConfig
-    from .models.mctct import MCTCT_PRETRAINED_CONFIG_ARCHIVE_MAP, MCTCTConfig, MCTCTFeatureExtractor, MCTCTProcessor
     from .models.mega import MEGA_PRETRAINED_CONFIG_ARCHIVE_MAP, MegaConfig
     from .models.megatron_bert import MEGATRON_BERT_PRETRAINED_CONFIG_ARCHIVE_MAP, MegatronBertConfig
     from .models.mgp_str import MGP_STR_PRETRAINED_CONFIG_ARCHIVE_MAP, MgpstrConfig, MgpstrProcessor, MgpstrTokenizer
-    from .models.mmbt import MMBTConfig
     from .models.mobilebert import MOBILEBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, MobileBertConfig, MobileBertTokenizer
     from .models.mobilenet_v1 import MOBILENET_V1_PRETRAINED_CONFIG_ARCHIVE_MAP, MobileNetV1Config
     from .models.mobilenet_v2 import MOBILENET_V2_PRETRAINED_CONFIG_ARCHIVE_MAP, MobileNetV2Config
@@ -4359,7 +4385,6 @@ if TYPE_CHECKING:
     from .models.regnet import REGNET_PRETRAINED_CONFIG_ARCHIVE_MAP, RegNetConfig
     from .models.rembert import REMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, RemBertConfig
     from .models.resnet import RESNET_PRETRAINED_CONFIG_ARCHIVE_MAP, ResNetConfig
-    from .models.retribert import RETRIBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, RetriBertConfig, RetriBertTokenizer
     from .models.roberta import ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, RobertaConfig, RobertaTokenizer
     from .models.roberta_prelayernorm import (
         ROBERTA_PRELAYERNORM_PRETRAINED_CONFIG_ARCHIVE_MAP,
@@ -4409,17 +4434,12 @@ if TYPE_CHECKING:
     from .models.t5 import T5_PRETRAINED_CONFIG_ARCHIVE_MAP, T5Config
     from .models.table_transformer import TABLE_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, TableTransformerConfig
     from .models.tapas import TAPAS_PRETRAINED_CONFIG_ARCHIVE_MAP, TapasConfig, TapasTokenizer
-    from .models.tapex import TapexTokenizer
     from .models.time_series_transformer import (
         TIME_SERIES_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP,
         TimeSeriesTransformerConfig,
     )
     from .models.timesformer import TIMESFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, TimesformerConfig
     from .models.timm_backbone import TimmBackboneConfig
-    from .models.trajectory_transformer import (
-        TRAJECTORY_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP,
-        TrajectoryTransformerConfig,
-    )
     from .models.transfo_xl import (
         TRANSFO_XL_PRETRAINED_CONFIG_ARCHIVE_MAP,
         TransfoXLConfig,
@@ -4432,7 +4452,6 @@ if TYPE_CHECKING:
     from .models.unispeech import UNISPEECH_PRETRAINED_CONFIG_ARCHIVE_MAP, UniSpeechConfig
     from .models.unispeech_sat import UNISPEECH_SAT_PRETRAINED_CONFIG_ARCHIVE_MAP, UniSpeechSatConfig
     from .models.upernet import UperNetConfig
-    from .models.van import VAN_PRETRAINED_CONFIG_ARCHIVE_MAP, VanConfig
     from .models.videomae import VIDEOMAE_PRETRAINED_CONFIG_ARCHIVE_MAP, VideoMAEConfig
     from .models.vilt import (
         VILT_PRETRAINED_CONFIG_ARCHIVE_MAP,
@@ -4667,6 +4686,7 @@ if TYPE_CHECKING:
         from .models.cpm import CpmTokenizerFast
         from .models.deberta import DebertaTokenizerFast
         from .models.deberta_v2 import DebertaV2TokenizerFast
+        from .models.deprecated.retribert import RetriBertTokenizerFast
         from .models.distilbert import DistilBertTokenizerFast
         from .models.dpr import DPRContextEncoderTokenizerFast, DPRQuestionEncoderTokenizerFast, DPRReaderTokenizerFast
         from .models.electra import ElectraTokenizerFast
@@ -4697,7 +4717,6 @@ if TYPE_CHECKING:
         from .models.realm import RealmTokenizerFast
         from .models.reformer import ReformerTokenizerFast
         from .models.rembert import RemBertTokenizerFast
-        from .models.retribert import RetriBertTokenizerFast
         from .models.roberta import RobertaTokenizerFast
         from .models.roformer import RoFormerTokenizerFast
         from .models.splinter import SplinterTokenizerFast
@@ -5262,6 +5281,29 @@ if TYPE_CHECKING:
             DeiTModel,
             DeiTPreTrainedModel,
         )
+        from .models.deprecated.mctct import (
+            MCTCT_PRETRAINED_MODEL_ARCHIVE_LIST,
+            MCTCTForCTC,
+            MCTCTModel,
+            MCTCTPreTrainedModel,
+        )
+        from .models.deprecated.mmbt import MMBTForClassification, MMBTModel, ModalEmbeddings
+        from .models.deprecated.retribert import (
+            RETRIBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
+            RetriBertModel,
+            RetriBertPreTrainedModel,
+        )
+        from .models.deprecated.trajectory_transformer import (
+            TRAJECTORY_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+            TrajectoryTransformerModel,
+            TrajectoryTransformerPreTrainedModel,
+        )
+        from .models.deprecated.van import (
+            VAN_PRETRAINED_MODEL_ARCHIVE_LIST,
+            VanForImageClassification,
+            VanModel,
+            VanPreTrainedModel,
+        )
         from .models.deta import (
             DETA_PRETRAINED_MODEL_ARCHIVE_LIST,
             DetaForObjectDetection,
@@ -5698,7 +5740,6 @@ if TYPE_CHECKING:
             MBartModel,
             MBartPreTrainedModel,
         )
-        from .models.mctct import MCTCT_PRETRAINED_MODEL_ARCHIVE_LIST, MCTCTForCTC, MCTCTModel, MCTCTPreTrainedModel
         from .models.mega import (
             MEGA_PRETRAINED_MODEL_ARCHIVE_LIST,
             MegaForCausalLM,
@@ -5729,7 +5770,6 @@ if TYPE_CHECKING:
             MgpstrModel,
             MgpstrPreTrainedModel,
         )
-        from .models.mmbt import MMBTForClassification, MMBTModel, ModalEmbeddings
         from .models.mobilebert import (
             MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
             MobileBertForMaskedLM,
@@ -6011,7 +6051,6 @@ if TYPE_CHECKING:
             ResNetModel,
             ResNetPreTrainedModel,
         )
-        from .models.retribert import RETRIBERT_PRETRAINED_MODEL_ARCHIVE_LIST, RetriBertModel, RetriBertPreTrainedModel
         from .models.roberta import (
             ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
             RobertaForCausalLM,
@@ -6204,11 +6243,6 @@ if TYPE_CHECKING:
             TimesformerPreTrainedModel,
         )
         from .models.timm_backbone import TimmBackbone
-        from .models.trajectory_transformer import (
-            TRAJECTORY_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
-            TrajectoryTransformerModel,
-            TrajectoryTransformerPreTrainedModel,
-        )
         from .models.transfo_xl import (
             TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST,
             AdaptiveEmbedding,
@@ -6252,12 +6286,6 @@ if TYPE_CHECKING:
             UniSpeechSatPreTrainedModel,
         )
         from .models.upernet import UperNetForSemanticSegmentation, UperNetPreTrainedModel
-        from .models.van import (
-            VAN_PRETRAINED_MODEL_ARCHIVE_LIST,
-            VanForImageClassification,
-            VanModel,
-            VanPreTrainedModel,
-        )
         from .models.videomae import (
             VIDEOMAE_PRETRAINED_MODEL_ARCHIVE_LIST,
             VideoMAEForPreTraining,

src/transformers/models/__init__.py

@@ -36,7 +36,6 @@ from . import (
     blip,
     blip_2,
     bloom,
-    bort,
     bridgetower,
     byt5,
     camembert,
@@ -60,6 +59,7 @@ from . import (
     decision_transformer,
     deformable_detr,
     deit,
+    deprecated,
     deta,
     detr,
     dialogpt,
@@ -122,13 +122,11 @@ from . import (
     maskformer,
     mbart,
     mbart50,
-    mctct,
     mega,
     megatron_bert,
     megatron_gpt2,
     mgp_str,
     mluke,
-    mmbt,
     mobilebert,
     mobilenet_v1,
     mobilenet_v2,
@@ -164,7 +162,6 @@ from . import (
     regnet,
     rembert,
     resnet,
-    retribert,
     roberta,
     roberta_prelayernorm,
     roc_bert,
@@ -188,11 +185,9 @@ from . import (
     t5,
     table_transformer,
     tapas,
-    tapex,
     time_series_transformer,
     timesformer,
     timm_backbone,
-    trajectory_transformer,
     transfo_xl,
     trocr,
     tvlt,
@@ -200,7 +195,6 @@ from . import (
     unispeech,
     unispeech_sat,
     upernet,
-    van,
     videomae,
     vilt,
     vision_encoder_decoder,

src/transformers/models/auto/configuration_auto.py

@@ -640,6 +640,15 @@ MODEL_NAMES_MAPPING = OrderedDict(
     ]
 )
 
+DEPRECATED_MODELS = [
+    "bort",
+    "mctct",
+    "mmbt",
+    "retribert",
+    "trajectory_transformer",
+    "van",
+]
+
 SPECIAL_MODEL_TYPE_TO_MODULE_NAME = OrderedDict(
     [
         ("openai-gpt", "openai"),
@@ -659,7 +668,11 @@ def model_type_to_module_name(key):
     if key in SPECIAL_MODEL_TYPE_TO_MODULE_NAME:
         return SPECIAL_MODEL_TYPE_TO_MODULE_NAME[key]
 
-    return key.replace("-", "_")
+    key = key.replace("-", "_")
+    if key in DEPRECATED_MODELS:
+        key = f"deprecated.{key}"
+
+    return key
 
 
 def config_class_to_model_type(config):

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

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

src/transformers/models/mctct/configuration_mctct.py → src/transformers/models/deprecated/mctct/configuration_mctct.py

@@ -14,8 +14,8 @@
 # limitations under the License.
 """M-CTC-T model configuration"""
 
-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/mctct/feature_extraction_mctct.py → src/transformers/models/deprecated/mctct/feature_extraction_mctct.py

@@ -20,11 +20,11 @@ from typing import List, Optional, Union
 
 import numpy as np
 
-from ...audio_utils import mel_filter_bank, optimal_fft_length, spectrogram, window_function
-from ...feature_extraction_sequence_utils import SequenceFeatureExtractor
-from ...feature_extraction_utils import BatchFeature
-from ...file_utils import PaddingStrategy, TensorType
-from ...utils import logging
+from ....audio_utils import mel_filter_bank, optimal_fft_length, spectrogram, window_function
+from ....feature_extraction_sequence_utils import SequenceFeatureExtractor
+from ....feature_extraction_utils import BatchFeature
+from ....file_utils import PaddingStrategy, TensorType
+from ....utils import logging
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/mctct/modeling_mctct.py → src/transformers/models/deprecated/mctct/modeling_mctct.py

@@ -22,17 +22,17 @@ import torch
 import torch.utils.checkpoint
 from torch import nn
 
-from ...activations import ACT2FN
-from ...deepspeed import is_deepspeed_zero3_enabled
-from ...file_utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward
-from ...modeling_outputs import BaseModelOutput, CausalLMOutput
-from ...modeling_utils import (
+from ....activations import ACT2FN
+from ....deepspeed import is_deepspeed_zero3_enabled
+from ....file_utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward
+from ....modeling_outputs import BaseModelOutput, CausalLMOutput
+from ....modeling_utils import (
     PreTrainedModel,
     apply_chunking_to_forward,
     find_pruneable_heads_and_indices,
     prune_linear_layer,
 )
-from ...utils import logging
+from ....utils import logging
 from .configuration_mctct import MCTCTConfig
 
 

src/transformers/models/mctct/processing_mctct.py → src/transformers/models/deprecated/mctct/processing_mctct.py

@@ -18,7 +18,7 @@ Speech processor class for M-CTC-T
 import warnings
 from contextlib import contextmanager
 
-from ...processing_utils import ProcessorMixin
+from ....processing_utils import ProcessorMixin
 
 
 class MCTCTProcessor(ProcessorMixin):

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

@@ -14,7 +14,7 @@
 
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
+from ....utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
 
 
 _import_structure = {"configuration_mmbt": ["MMBTConfig"]}

src/transformers/models/mmbt/configuration_mmbt.py → src/transformers/models/deprecated/mmbt/configuration_mmbt.py

@@ -15,7 +15,7 @@
 # limitations under the License.
 """ MMBT configuration"""
 
-from ...utils import logging
+from ....utils import logging
 
 
 logger = logging.get_logger(__name__)

src/transformers/models/mmbt/modeling_mmbt.py → src/transformers/models/deprecated/mmbt/modeling_mmbt.py

@@ -20,9 +20,9 @@ import torch
 from torch import nn
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from ...modeling_outputs import BaseModelOutputWithPooling, SequenceClassifierOutput
-from ...modeling_utils import ModuleUtilsMixin
-from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
+from ....modeling_outputs import BaseModelOutputWithPooling, SequenceClassifierOutput
+from ....modeling_utils import ModuleUtilsMixin
+from ....utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
 
 
 logger = logging.get_logger(__name__)