Revert "Add Dia model (huggingface#38405)"

This reverts commit 583db52.

# Conflicts:
#	docs/source/en/model_doc/dia.md
#	src/transformers/configuration_utils.py
#	src/transformers/models/auto/processing_auto.py
#	src/transformers/models/auto/tokenization_auto.py
#	src/transformers/models/dac/modeling_dac.py
#	src/transformers/models/dia/generation_dia.py
#	src/transformers/models/dia/modeling_dia.py
#	src/transformers/models/dia/modular_dia.py
#	tests/models/dia/test_feature_extraction_dia.py
#	tests/models/dia/test_modeling_dia.py
#	tests/models/dia/test_processing_dia.py
#	utils/check_config_attributes.py

Author: zaristei

docs/source/en/_toctree.yml

@@ -879,8 +879,6 @@
         title: CSM
       - local: model_doc/dac
         title: dac
-      - local: model_doc/dia
-        title: Dia
       - local: model_doc/encodec
         title: EnCodec
       - local: model_doc/fastspeech2_conformer

docs/source/en/model_doc/auto.md

@@ -354,10 +354,6 @@ The following auto classes are available for the following audio tasks.
 
 [[autodoc]] AutoModelForTextToWaveform
 
-### AutoModelForAudioTokenization
-
-[[autodoc]] AutoModelForAudioTokenization
-
 ## Multimodal
 
 The following auto classes are available for the following multimodal tasks.

docs/source/en/model_doc/dia.md

@@ -319,6 +319,64 @@ def __init__(
         for parameter_name, default_value in self._get_global_generation_defaults().items():
             setattr(self, parameter_name, kwargs.pop(parameter_name, default_value))
 
+        # Fine-tuning task arguments
+        self.architectures = kwargs.pop("architectures", None)
+        self.finetuning_task = kwargs.pop("finetuning_task", None)
+        self.id2label = kwargs.pop("id2label", None)
+        self.label2id = kwargs.pop("label2id", None)
+        if self.label2id is not None and not isinstance(self.label2id, dict):
+            raise ValueError("Argument label2id should be a dictionary.")
+        if self.id2label is not None:
+            if not isinstance(self.id2label, dict):
+                raise ValueError("Argument id2label should be a dictionary.")
+            num_labels = kwargs.pop("num_labels", None)
+            if num_labels is not None and len(self.id2label) != num_labels:
+                logger.warning(
+                    f"You passed along `num_labels={num_labels}` with an incompatible id to label map: "
+                    f"{self.id2label}. The number of labels will be overwritten to {self.num_labels}."
+                )
+            self.id2label = {int(key): value for key, value in self.id2label.items()}
+            # Keys are always strings in JSON so convert ids to int here.
+        else:
+            self.num_labels = kwargs.pop("num_labels", 2)
+
+        if self.torch_dtype is not None and isinstance(self.torch_dtype, str):
+            # we will start using self.torch_dtype in v5, but to be consistent with
+            # from_pretrained's torch_dtype arg convert it to an actual torch.dtype object
+            if is_torch_available():
+                import torch
+
+                self.torch_dtype = getattr(torch, self.torch_dtype)
+
+        # Tokenizer arguments TODO: eventually tokenizer and models should share the same config
+        self.tokenizer_class = kwargs.pop("tokenizer_class", None)
+        self.prefix = kwargs.pop("prefix", None)
+        self.bos_token_id = kwargs.pop("bos_token_id", None)
+        self.pad_token_id = kwargs.pop("pad_token_id", None)
+        self.eos_token_id = kwargs.pop("eos_token_id", None)
+        self.sep_token_id = kwargs.pop("sep_token_id", None)
+
+        self.decoder_start_token_id = kwargs.pop("decoder_start_token_id", None)
+
+        # task specific arguments
+        self.task_specific_params = kwargs.pop("task_specific_params", None)
+
+        # regression / multi-label classification
+        self.problem_type = kwargs.pop("problem_type", None)
+        allowed_problem_types = ("regression", "single_label_classification", "multi_label_classification")
+        if self.problem_type is not None and self.problem_type not in allowed_problem_types:
+            raise ValueError(
+                f"The config parameter `problem_type` was not understood: received {self.problem_type} "
+                "but only 'regression', 'single_label_classification' and 'multi_label_classification' are valid."
+            )
+
+        # TPU arguments
+        if kwargs.pop("xla_device", None) is not None:
+            logger.warning(
+                "The `xla_device` argument has been deprecated in v4.4.0 of Transformers. It is ignored and you can "
+                "safely remove it from your `config.json` file."
+            )
+
         # Name or path to the pretrained checkpoint
         self._name_or_path = str(kwargs.pop("name_or_path", ""))
         self._commit_hash = kwargs.pop("_commit_hash", None)