# This file is autogenerated by the command `make fix-copies`, do not edit.
from ..file_utils import requires_backends


class PyTorchBenchmark:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PyTorchBenchmarkArguments:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GlueDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GlueDataTrainingArguments:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LineByLineTextDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LineByLineWithRefDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LineByLineWithSOPTextDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SquadDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SquadDataTrainingArguments:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TextDataset:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TextDatasetForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeamScorer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeamSearchScorer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ForcedBOSTokenLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class ForcedEOSTokenLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class HammingDiversityLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class InfNanRemoveLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class LogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class LogitsProcessorList:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class LogitsWarper:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MinLengthLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class NoBadWordsLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class NoRepeatNGramLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class PrefixConstrainedLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class RepetitionPenaltyLogitsProcessor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])


class TemperatureLogitsWarper:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TopKLogitsWarper:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TopPLogitsWarper:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MaxLengthCriteria:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MaxTimeCriteria:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class StoppingCriteria:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class StoppingCriteriaList:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def top_k_top_p_filtering(*args, **kwargs):
    requires_backends(top_k_top_p_filtering, ["torch"])


class Conv1D:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def apply_chunking_to_forward(*args, **kwargs):
    requires_backends(apply_chunking_to_forward, ["torch"])


def prune_layer(*args, **kwargs):
    requires_backends(prune_layer, ["torch"])


ALBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class AlbertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AlbertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_albert(*args, **kwargs):
    requires_backends(load_tf_weights_in_albert, ["torch"])


MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING = None


MODEL_FOR_CAUSAL_IMAGE_MODELING_MAPPING = None


MODEL_FOR_CAUSAL_LM_MAPPING = None


MODEL_FOR_CTC_MAPPING = None


MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING = None


MODEL_FOR_IMAGE_SEGMENTATION_MAPPING = None


MODEL_FOR_MASKED_LM_MAPPING = None


MODEL_FOR_MULTIPLE_CHOICE_MAPPING = None


MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING = None


MODEL_FOR_OBJECT_DETECTION_MAPPING = None


MODEL_FOR_PRETRAINING_MAPPING = None


MODEL_FOR_QUESTION_ANSWERING_MAPPING = None


MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING = None


MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING = None


MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING = None


MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING = None


MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING = None


MODEL_FOR_VISION_2_SEQ_MAPPING = None


MODEL_MAPPING = None


MODEL_WITH_LM_HEAD_MAPPING = None


class AutoModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForAudioClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForAudioFrameClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForAudioXVector:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForImageSegmentation:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForObjectDetection:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForSeq2SeqLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForSpeechSeq2Seq:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForTableQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelForVision2Seq:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AutoModelWithLMHead:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


BART_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BartForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BartForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BartForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BartForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BartModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BartPretrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PretrainedBartModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


BEIT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BeitForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeitForMaskedImageModeling:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeitForSemanticSegmentation:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeitModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BeitPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


BERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_bert(*args, **kwargs):
    requires_backends(load_tf_weights_in_bert, ["torch"])


class BertGenerationDecoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertGenerationEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BertGenerationPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_bert_generation(*args, **kwargs):
    requires_backends(load_tf_weights_in_bert_generation, ["torch"])


BIG_BIRD_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BigBirdForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_big_bird(*args, **kwargs):
    requires_backends(load_tf_weights_in_big_bird, ["torch"])


BIGBIRD_PEGASUS_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BigBirdPegasusForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPegasusForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPegasusForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPegasusForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPegasusModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BigBirdPegasusPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


BLENDERBOT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BlenderbotForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


BLENDERBOT_SMALL_PRETRAINED_MODEL_ARCHIVE_LIST = None


class BlenderbotSmallForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotSmallForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotSmallModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class BlenderbotSmallPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class CamembertForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CamembertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


CANINE_PRETRAINED_MODEL_ARCHIVE_LIST = None


class CanineForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CanineForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CanineForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CanineForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CanineLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CanineModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CaninePreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_canine(*args, **kwargs):
    requires_backends(load_tf_weights_in_canine, ["torch"])


CLIP_PRETRAINED_MODEL_ARCHIVE_LIST = None


class CLIPModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CLIPPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CLIPTextModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CLIPVisionModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ConvBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ConvBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_convbert(*args, **kwargs):
    requires_backends(load_tf_weights_in_convbert, ["torch"])


CTRL_PRETRAINED_MODEL_ARCHIVE_LIST = None


class CTRLForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CTRLLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CTRLModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class CTRLPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST = None


class DebertaForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


DEBERTA_V2_PRETRAINED_MODEL_ARCHIVE_LIST = None


class DebertaV2ForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaV2ForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaV2ForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaV2ForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaV2Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DebertaV2PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


DEIT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class DeiTForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DeiTForImageClassificationWithTeacher:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DeiTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DeiTPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


DISTILBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class DistilBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DistilBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


DPR_CONTEXT_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST = None


DPR_QUESTION_ENCODER_PRETRAINED_MODEL_ARCHIVE_LIST = None


DPR_READER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class DPRContextEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRPretrainedContextEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRPretrainedQuestionEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRPretrainedReader:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRQuestionEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class DPRReader:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


ELECTRA_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ElectraForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ElectraPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_electra(*args, **kwargs):
    requires_backends(load_tf_weights_in_electra, ["torch"])


class EncoderDecoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


FLAUBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class FlaubertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertForQuestionAnsweringSimple:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FlaubertWithLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


FNET_PRETRAINED_MODEL_ARCHIVE_LIST = None


class FNetForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FNetPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FSMTForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FSMTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PretrainedFSMTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


FUNNEL_PRETRAINED_MODEL_ARCHIVE_LIST = None


class FunnelBaseModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class FunnelPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_funnel(*args, **kwargs):
    requires_backends(load_tf_weights_in_funnel, ["torch"])


GPT2_PRETRAINED_MODEL_ARCHIVE_LIST = None


class GPT2DoubleHeadsModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPT2ForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPT2ForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPT2LMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPT2Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPT2PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_gpt2(*args, **kwargs):
    requires_backends(load_tf_weights_in_gpt2, ["torch"])


GPT_NEO_PRETRAINED_MODEL_ARCHIVE_LIST = None


class GPTNeoForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTNeoForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTNeoModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTNeoPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_gpt_neo(*args, **kwargs):
    requires_backends(load_tf_weights_in_gpt_neo, ["torch"])


GPTJ_PRETRAINED_MODEL_ARCHIVE_LIST = None


class GPTJForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTJForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTJForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTJModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class GPTJPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


HUBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class HubertForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class HubertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class HubertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class HubertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


IBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class IBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class IBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


IMAGEGPT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ImageGPTForCausalImageModeling:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ImageGPTForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ImageGPTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ImageGPTPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_imagegpt(*args, **kwargs):
    requires_backends(load_tf_weights_in_imagegpt, ["torch"])


LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST = None


class LayoutLMForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


LAYOUTLMV2_PRETRAINED_MODEL_ARCHIVE_LIST = None


class LayoutLMv2ForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMv2ForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMv2ForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMv2Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LayoutLMv2PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


LED_PRETRAINED_MODEL_ARCHIVE_LIST = None


class LEDForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LEDForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LEDForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LEDModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LEDPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class LongformerForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LongformerSelfAttention:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


LUKE_PRETRAINED_MODEL_ARCHIVE_LIST = None


class LukeForEntityClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LukeForEntityPairClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LukeForEntitySpanClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LukeForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LukeModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LukePreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertVisualFeatureEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class LxmertXLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


M2M_100_PRETRAINED_MODEL_ARCHIVE_LIST = None


class M2M100ForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class M2M100Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class M2M100PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MarianForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MarianModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MarianMTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MBartPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


MEGATRON_BERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class MegatronBertForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MegatronBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MMBTForClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MMBTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ModalEmbeddings:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class MobileBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForNextSentencePrediction:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MobileBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_mobilebert(*args, **kwargs):
    requires_backends(load_tf_weights_in_mobilebert, ["torch"])


MPNET_PRETRAINED_MODEL_ARCHIVE_LIST = None


class MPNetForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MPNetPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MT5EncoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MT5ForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class MT5Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class OpenAIGPTDoubleHeadsModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class OpenAIGPTForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class OpenAIGPTLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class OpenAIGPTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class OpenAIGPTPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_openai_gpt(*args, **kwargs):
    requires_backends(load_tf_weights_in_openai_gpt, ["torch"])


class PegasusForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PegasusForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PegasusModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PegasusPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


PERCEIVER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class PerceiverForImageClassificationConvProcessing:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForImageClassificationFourier:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForImageClassificationLearned:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForMultimodalAutoencoding:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForOpticalFlow:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class PerceiverPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


PROPHETNET_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ProphetNetDecoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ProphetNetEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ProphetNetForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ProphetNetForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ProphetNetModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ProphetNetPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RagModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RagPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RagSequenceForGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RagTokenForGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


REFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ReformerAttention:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerModelWithLMHead:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ReformerPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


REMBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class RemBertForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RemBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_rembert(*args, **kwargs):
    requires_backends(load_tf_weights_in_rembert, ["torch"])


RETRIBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class RetriBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RetriBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST = None


class RobertaForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RobertaPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class RoFormerForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class RoFormerPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_roformer(*args, **kwargs):
    requires_backends(load_tf_weights_in_roformer, ["torch"])


SEGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class SegformerDecodeHead:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SegformerForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SegformerForSemanticSegmentation:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SegformerLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SegformerModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SegformerPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


SEW_PRETRAINED_MODEL_ARCHIVE_LIST = None


class SEWForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


SEW_D_PRETRAINED_MODEL_ARCHIVE_LIST = None


class SEWDForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWDForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWDModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SEWDPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SpeechEncoderDecoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


SPEECH_TO_TEXT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class Speech2TextForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Speech2TextModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Speech2TextPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Speech2Text2ForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Speech2Text2PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


SPLINTER_PRETRAINED_MODEL_ARCHIVE_LIST = None


class SplinterForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SplinterLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SplinterModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SplinterPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


SQUEEZEBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class SqueezeBertForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertModule:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class SqueezeBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


T5_PRETRAINED_MODEL_ARCHIVE_LIST = None


class T5EncoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class T5ForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class T5Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class T5PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_t5(*args, **kwargs):
    requires_backends(load_tf_weights_in_t5, ["torch"])


TRANSFO_XL_PRETRAINED_MODEL_ARCHIVE_LIST = None


class AdaptiveEmbedding:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TransfoXLForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TransfoXLLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TransfoXLModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TransfoXLPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_transfo_xl(*args, **kwargs):
    requires_backends(load_tf_weights_in_transfo_xl, ["torch"])


TROCR_PRETRAINED_MODEL_ARCHIVE_LIST = None


class TrOCRForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class TrOCRPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


UNISPEECH_PRETRAINED_MODEL_ARCHIVE_LIST = None


class UniSpeechForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


UNISPEECH_SAT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class UniSpeechSatForAudioFrameClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatForXVector:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class UniSpeechSatPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisionEncoderDecoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisionTextDualEncoderModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


VISUAL_BERT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class VisualBertForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertForRegionToPhraseAlignment:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertForVisualReasoning:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertLayer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class VisualBertPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


VIT_PRETRAINED_MODEL_ARCHIVE_LIST = None


class ViTForImageClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ViTModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class ViTPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST = None


class Wav2Vec2ForAudioFrameClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2ForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2ForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2ForPreTraining:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2ForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2ForXVector:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2Model:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class Wav2Vec2PreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


WAVLM_PRETRAINED_MODEL_ARCHIVE_LIST = None


class WavLMForAudioFrameClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class WavLMForCTC:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class WavLMForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class WavLMForXVector:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class WavLMModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class WavLMPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


XLM_PRETRAINED_MODEL_ARCHIVE_LIST = None


class XLMForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMForQuestionAnsweringSimple:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMWithLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


XLM_PROPHETNET_PRETRAINED_MODEL_ARCHIVE_LIST = None


class XLMProphetNetDecoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMProphetNetEncoder:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMProphetNetForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMProphetNetForConditionalGeneration:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMProphetNetModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST = None


class XLMRobertaForCausalLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaForMaskedLM:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLMRobertaModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


XLNET_PRETRAINED_MODEL_ARCHIVE_LIST = None


class XLNetForMultipleChoice:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetForQuestionAnswering:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetForQuestionAnsweringSimple:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetForSequenceClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetForTokenClassification:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetLMHeadModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class XLNetPreTrainedModel:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        requires_backends(cls, ["torch"])

    def forward(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def load_tf_weights_in_xlnet(*args, **kwargs):
    requires_backends(load_tf_weights_in_xlnet, ["torch"])


class Adafactor:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


class AdamW:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def get_constant_schedule(*args, **kwargs):
    requires_backends(get_constant_schedule, ["torch"])


def get_constant_schedule_with_warmup(*args, **kwargs):
    requires_backends(get_constant_schedule_with_warmup, ["torch"])


def get_cosine_schedule_with_warmup(*args, **kwargs):
    requires_backends(get_cosine_schedule_with_warmup, ["torch"])


def get_cosine_with_hard_restarts_schedule_with_warmup(*args, **kwargs):
    requires_backends(get_cosine_with_hard_restarts_schedule_with_warmup, ["torch"])


def get_linear_schedule_with_warmup(*args, **kwargs):
    requires_backends(get_linear_schedule_with_warmup, ["torch"])


def get_polynomial_decay_schedule_with_warmup(*args, **kwargs):
    requires_backends(get_polynomial_decay_schedule_with_warmup, ["torch"])


def get_scheduler(*args, **kwargs):
    requires_backends(get_scheduler, ["torch"])


class Trainer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])


def torch_distributed_zero_first(*args, **kwargs):
    requires_backends(torch_distributed_zero_first, ["torch"])


class Seq2SeqTrainer:
    def __init__(self, *args, **kwargs):
        requires_backends(self, ["torch"])