# DeBERTa[[deberta]] ## 개요[[overview]] DeBERTa 모델은 Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen이 작성한 [DeBERTa: 분리된 어텐션을 활용한 디코딩 강화 BERT](https://huggingface.co/papers/2006.03654)이라는 논문에서 제안되었습니다. 이 모델은 2018년 Google이 발표한 BERT 모델과 2019년 Facebook이 발표한 RoBERTa 모델을 기반으로 합니다. DeBERTa는 RoBERTa에서 사용된 데이터의 절반만을 사용하여 분리된(disentangled) 어텐션과 향상된 마스크 디코더 학습을 통해 RoBERTa를 개선했습니다. 논문의 초록은 다음과 같습니다: *사전 학습된 신경망 언어 모델의 최근 발전은 많은 자연어 처리(NLP) 작업의 성능을 크게 향상시켰습니다. 본 논문에서는 두 가지 새로운 기술을 사용하여 BERT와 RoBERTa 모델을 개선한 새로운 모델 구조인 DeBERTa를 제안합니다. 첫 번째는 분리된 어텐션 메커니즘으로, 각 단어가 내용과 위치를 각각 인코딩하는 두 개의 벡터로 표현되며, 단어들 간의 어텐션 가중치는 내용과 상대적 위치에 대한 분리된 행렬을 사용하여 계산됩니다. 두 번째로, 모델 사전 학습을 위해 마스킹된 토큰을 예측하는 출력 소프트맥스 층을 대체하는 향상된 마스크 디코더가 사용됩니다. 우리는 이 두 가지 기술이 모델 사전 학습의 효율성과 다운스트림 작업의 성능을 크게 향상시킨다는 것을 보여줍니다. RoBERTa-Large와 비교했을 때, 절반의 학습 데이터로 학습된 DeBERTa 모델은 광범위한 NLP 작업에서 일관되게 더 나은 성능을 보여주며, MNLI에서 +0.9%(90.2% vs 91.1%), SQuAD v2.0에서 +2.3%(88.4% vs 90.7%), RACE에서 +3.6%(83.2% vs 86.8%)의 성능 향상을 달성했습니다. DeBERTa 코드와 사전 학습된 모델은 https://github.com/microsoft/DeBERTa 에서 공개될 예정입니다.* [DeBERTa](https://huggingface.co/DeBERTa) 모델의 텐서플로 2.0 구현은 [kamalkraj](https://huggingface.co/kamalkraj)가 기여했습니다. 원본 코드는 [이곳](https://github.com/microsoft/DeBERTa)에서 확인하실 수 있습니다. ## 리소스[[resources]] DeBERTa를 시작하는 데 도움이 되는 Hugging Face와 community 자료 목록(🌎로 표시됨) 입니다. 여기에 포함될 자료를 제출하고 싶으시다면 PR(Pull Request)를 열어주세요. 리뷰해 드리겠습니다! 자료는 기존 자료를 복제하는 대신 새로운 내용을 담고 있어야 합니다. - DeBERTa와 [DeepSpeed를 이용해서 대형 모델 학습을 가속시키는](https://huggingface.co/blog/accelerate-deepspeed) 방법에 대한 포스트. - DeBERTa와 [머신러닝으로 한층 향상된 고객 서비스](https://huggingface.co/blog/supercharge-customer-service-with-machine-learning)에 대한 블로그 포스트. - [DebertaForSequenceClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForSequenceClassification)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/pytorch/text-classification)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb)에서 지원됩니다. - [TFDebertaForSequenceClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForSequenceClassification)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/text-classification)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb)에서 지원됩니다. - [텍스트 분류 작업 가이드](../tasks/sequence_classification) - [DebertaForTokenClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForTokenClassification)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/pytorch/token-classification)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb)에서 지원합니다. - [TFDebertaForTokenClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForTokenClassification)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/token-classification)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb)에서 지원합니다. - 🤗 Hugging Face 코스의 [토큰 분류](https://huggingface.co/course/chapter7/2?fw=pt) 장. - 🤗 Hugging Face 코스의 [BPE(Byte-Pair Encoding) 토큰화](https://huggingface.co/course/chapter6/5?fw=pt) 장. - [토큰 분류 작업 가이드](../tasks/token_classification) - [DebertaForMaskedLM](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForMaskedLM)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/pytorch/language-modeling#robertabertdistilbert-and-masked-language-modeling)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling.ipynb)에서 지원합니다. - [TFDebertaForMaskedLM](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForMaskedLM)은 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/language-modeling#run_mlmpy)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb)에서 지원합니다. - 🤗 Hugging Face 코스의 [마스크 언어 모델링](https://huggingface.co/course/chapter7/3?fw=pt) 장. - [마스크 언어 모델링 작업 가이드](../tasks/masked_language_modeling) - [DebertaForQuestionAnswering](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForQuestionAnswering)은 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb)에서 지원합니다. - [TFDebertaForQuestionAnswering](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForQuestionAnswering)는 이 [예제 스크립트](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/question-answering)와 [노트북](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb)에서 지원합니다. - 🤗 Hugging Face 코스의 [질의응답(Question answering)](https://huggingface.co/course/chapter7/7?fw=pt) 장. - [질의응답 작업 가이드](../tasks/question_answering) ## DebertaConfig[[transformers.DebertaConfig]][[transformers.DebertaConfig]] #### transformers.DebertaConfig[[transformers.DebertaConfig]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/configuration_deberta.py#L33) This is the configuration class to store the configuration of a [DebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaModel) or a [TFDebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaModel). It is used to instantiate a DeBERTa model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the DeBERTa [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base) architecture. Configuration objects inherit from [PretrainedConfig](/docs/transformers/v4.57.1/ko/main_classes/configuration#transformers.PretrainedConfig) and can be used to control the model outputs. Read the documentation from [PretrainedConfig](/docs/transformers/v4.57.1/ko/main_classes/configuration#transformers.PretrainedConfig) for more information. Example: ```python >>> from transformers import DebertaConfig, DebertaModel >>> # Initializing a DeBERTa microsoft/deberta-base style configuration >>> configuration = DebertaConfig() >>> # Initializing a model (with random weights) from the microsoft/deberta-base style configuration >>> model = DebertaModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` **Parameters:** vocab_size (`int`, *optional*, defaults to 50265) : Vocabulary size of the DeBERTa model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [DebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaModel) or [TFDebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaModel). hidden_size (`int`, *optional*, defaults to 768) : Dimensionality of the encoder layers and the pooler layer. num_hidden_layers (`int`, *optional*, defaults to 12) : Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 12) : Number of attention heads for each attention layer in the Transformer encoder. intermediate_size (`int`, *optional*, defaults to 3072) : Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder. hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`) : The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"silu"`, `"gelu"`, `"tanh"`, `"gelu_fast"`, `"mish"`, `"linear"`, `"sigmoid"` and `"gelu_new"` are supported. hidden_dropout_prob (`float`, *optional*, defaults to 0.1) : The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1) : The dropout ratio for the attention probabilities. max_position_embeddings (`int`, *optional*, defaults to 512) : The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048). type_vocab_size (`int`, *optional*, defaults to 0) : The vocabulary size of the `token_type_ids` passed when calling [DebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaModel) or [TFDebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaModel). initializer_range (`float`, *optional*, defaults to 0.02) : The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to 1e-12) : The epsilon used by the layer normalization layers. relative_attention (`bool`, *optional*, defaults to `False`) : Whether use relative position encoding. max_relative_positions (`int`, *optional*, defaults to 1) : The range of relative positions `[-max_position_embeddings, max_position_embeddings]`. Use the same value as `max_position_embeddings`. pad_token_id (`int`, *optional*, defaults to 0) : The value used to pad input_ids. position_biased_input (`bool`, *optional*, defaults to `True`) : Whether add absolute position embedding to content embedding. pos_att_type (`list[str]`, *optional*) : The type of relative position attention, it can be a combination of `["p2c", "c2p"]`, e.g. `["p2c"]`, `["p2c", "c2p"]`. layer_norm_eps (`float`, *optional*, defaults to 1e-12) : The epsilon used by the layer normalization layers. legacy (`bool`, *optional*, defaults to `True`) : Whether or not the model should use the legacy `LegacyDebertaOnlyMLMHead`, which does not work properly for mask infilling tasks. ## DebertaTokenizer[[transformers.DebertaTokenizer]][[transformers.DebertaTokenizer]] #### transformers.DebertaTokenizer[[transformers.DebertaTokenizer]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta.py#L72) Construct a DeBERTa tokenizer. Based on byte-level Byte-Pair-Encoding. This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import DebertaTokenizer >>> tokenizer = DebertaTokenizer.from_pretrained("microsoft/deberta-base") >>> tokenizer("Hello world")["input_ids"] [1, 31414, 232, 2] >>> tokenizer(" Hello world")["input_ids"] [1, 20920, 232, 2] ``` You can get around that behavior by passing `add_prefix_space=True` when instantiating this tokenizer or when you call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance. When used with `is_split_into_words=True`, this tokenizer will add a space before each word (even the first one). This tokenizer inherits from [PreTrainedTokenizer](/docs/transformers/v4.57.1/ko/main_classes/tokenizer#transformers.PreTrainedTokenizer) which contains most of the main methods. Users should refer to this superclass for more information regarding those methods. build_inputs_with_special_tokenstransformers.DebertaTokenizer.build_inputs_with_special_tokenshttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta.py#L249[{"name": "token_ids_0", "val": ": list"}, {"name": "token_ids_1", "val": ": typing.Optional[list[int]] = None"}]- **token_ids_0** (`List[int]`) -- List of IDs to which the special tokens will be added. - **token_ids_1** (`List[int]`, *optional*) -- Optional second list of IDs for sequence pairs.0`List[int]`List of [input IDs](../glossary#input-ids) with the appropriate special tokens. Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. A DeBERTa sequence has the following format: - single sequence: [CLS] X [SEP] - pair of sequences: [CLS] A [SEP] B [SEP] **Parameters:** vocab_file (`str`) : Path to the vocabulary file. merges_file (`str`) : Path to the merges file. errors (`str`, *optional*, defaults to `"replace"`) : Paradigm to follow when decoding bytes to UTF-8. See [bytes.decode](https://docs.python.org/3/library/stdtypes.html#bytes.decode) for more information. bos_token (`str`, *optional*, defaults to `"[CLS]"`) : The beginning of sequence token. eos_token (`str`, *optional*, defaults to `"[SEP]"`) : The end of sequence token. sep_token (`str`, *optional*, defaults to `"[SEP]"`) : The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens. cls_token (`str`, *optional*, defaults to `"[CLS]"`) : The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens. unk_token (`str`, *optional*, defaults to `"[UNK]"`) : The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead. pad_token (`str`, *optional*, defaults to `"[PAD]"`) : The token used for padding, for example when batching sequences of different lengths. mask_token (`str`, *optional*, defaults to `"[MASK]"`) : The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict. add_prefix_space (`bool`, *optional*, defaults to `False`) : Whether or not to add an initial space to the input. This allows to treat the leading word just as any other word. (Deberta tokenizer detect beginning of words by the preceding space). add_bos_token (`bool`, *optional*, defaults to `False`) : Whether or not to add an initial to the input. This allows to treat the leading word just as any other word. **Returns:** ``List[int]`` List of [input IDs](../glossary#input-ids) with the appropriate special tokens. #### get_special_tokens_mask[[transformers.DebertaTokenizer.get_special_tokens_mask]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta.py#L274) Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding special tokens using the tokenizer `prepare_for_model` or `encode_plus` methods. **Parameters:** token_ids_0 (`List[int]`) : List of IDs. token_ids_1 (`List[int]`, *optional*) : Optional second list of IDs for sequence pairs. already_has_special_tokens (`bool`, *optional*, defaults to `False`) : Whether or not the token list is already formatted with special tokens for the model. **Returns:** ``List[int]`` A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token. #### create_token_type_ids_from_sequences[[transformers.DebertaTokenizer.create_token_type_ids_from_sequences]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/tokenization_utils_base.py#L3543) Create the token type IDs corresponding to the sequences passed. [What are token type IDs?](../glossary#token-type-ids) Should be overridden in a subclass if the model has a special way of building those. **Parameters:** token_ids_0 (`list[int]`) : The first tokenized sequence. token_ids_1 (`list[int]`, *optional*) : The second tokenized sequence. **Returns:** ``list[int]`` The token type ids. #### save_vocabulary[[transformers.DebertaTokenizer.save_vocabulary]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta.py#L330) ## DebertaTokenizerFast[[transformers.DebertaTokenizerFast]][[transformers.DebertaTokenizerFast]] #### transformers.DebertaTokenizerFast[[transformers.DebertaTokenizerFast]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta_fast.py#L30) Construct a "fast" DeBERTa tokenizer (backed by HuggingFace's *tokenizers* library). Based on byte-level Byte-Pair-Encoding. This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import DebertaTokenizerFast >>> tokenizer = DebertaTokenizerFast.from_pretrained("microsoft/deberta-base") >>> tokenizer("Hello world")["input_ids"] [1, 31414, 232, 2] >>> tokenizer(" Hello world")["input_ids"] [1, 20920, 232, 2] ``` You can get around that behavior by passing `add_prefix_space=True` when instantiating this tokenizer, but since the model was not pretrained this way, it might yield a decrease in performance. When used with `is_split_into_words=True`, this tokenizer needs to be instantiated with `add_prefix_space=True`. This tokenizer inherits from [PreTrainedTokenizerFast](/docs/transformers/v4.57.1/ko/main_classes/tokenizer#transformers.PreTrainedTokenizerFast) which contains most of the main methods. Users should refer to this superclass for more information regarding those methods. build_inputs_with_special_tokenstransformers.DebertaTokenizerFast.build_inputs_with_special_tokenshttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/tokenization_deberta_fast.py#L157[{"name": "token_ids_0", "val": ": list"}, {"name": "token_ids_1", "val": ": typing.Optional[list[int]] = None"}]- **token_ids_0** (`List[int]`) -- List of IDs to which the special tokens will be added. - **token_ids_1** (`List[int]`, *optional*) -- Optional second list of IDs for sequence pairs.0`List[int]`List of [input IDs](../glossary#input-ids) with the appropriate special tokens. Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. A DeBERTa sequence has the following format: - single sequence: [CLS] X [SEP] - pair of sequences: [CLS] A [SEP] B [SEP] **Parameters:** vocab_file (`str`, *optional*) : Path to the vocabulary file. merges_file (`str`, *optional*) : Path to the merges file. tokenizer_file (`str`, *optional*) : The path to a tokenizer file to use instead of the vocab file. errors (`str`, *optional*, defaults to `"replace"`) : Paradigm to follow when decoding bytes to UTF-8. See [bytes.decode](https://docs.python.org/3/library/stdtypes.html#bytes.decode) for more information. bos_token (`str`, *optional*, defaults to `"[CLS]"`) : The beginning of sequence token. eos_token (`str`, *optional*, defaults to `"[SEP]"`) : The end of sequence token. sep_token (`str`, *optional*, defaults to `"[SEP]"`) : The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens. cls_token (`str`, *optional*, defaults to `"[CLS]"`) : The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens. unk_token (`str`, *optional*, defaults to `"[UNK]"`) : The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead. pad_token (`str`, *optional*, defaults to `"[PAD]"`) : The token used for padding, for example when batching sequences of different lengths. mask_token (`str`, *optional*, defaults to `"[MASK]"`) : The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict. add_prefix_space (`bool`, *optional*, defaults to `False`) : Whether or not to add an initial space to the input. This allows to treat the leading word just as any other word. (Deberta tokenizer detect beginning of words by the preceding space). **Returns:** ``List[int]`` List of [input IDs](../glossary#input-ids) with the appropriate special tokens. #### create_token_type_ids_from_sequences[[transformers.DebertaTokenizerFast.create_token_type_ids_from_sequences]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/tokenization_utils_base.py#L3543) Create the token type IDs corresponding to the sequences passed. [What are token type IDs?](../glossary#token-type-ids) Should be overridden in a subclass if the model has a special way of building those. **Parameters:** token_ids_0 (`list[int]`) : The first tokenized sequence. token_ids_1 (`list[int]`, *optional*) : The second tokenized sequence. **Returns:** ``list[int]`` The token type ids. ## DebertaModel[[transformers.DebertaModel]][[transformers.DebertaModel]] #### transformers.DebertaModel[[transformers.DebertaModel]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L640) The bare Deberta Model outputting raw hidden-states without any specific head on top. This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.DebertaModel.forwardhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L664[{"name": "input_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "token_type_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[transformers.modeling_outputs.BaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.BaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model. - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [DebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaModel) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. **Parameters:** config ([DebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaModel)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_outputs.BaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutput) or `tuple(torch.FloatTensor)`` A [transformers.modeling_outputs.BaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model. - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## DebertaPreTrainedModel[[transformers.DebertaPreTrainedModel]][[transformers.DebertaPreTrainedModel]] #### transformers.DebertaPreTrainedModel[[transformers.DebertaPreTrainedModel]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L611) This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. **Parameters:** config ([PretrainedConfig](/docs/transformers/v4.57.1/ko/main_classes/configuration#transformers.PretrainedConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. ## DebertaForMaskedLM[[transformers.DebertaForMaskedLM]][[transformers.DebertaForMaskedLM]] #### transformers.DebertaForMaskedLM[[transformers.DebertaForMaskedLM]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L839) The Deberta Model with a `language modeling` head on top." This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.DebertaForMaskedLM.forwardhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L869[{"name": "input_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "token_type_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "labels", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **labels** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ..., config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]` - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[transformers.modeling_outputs.MaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.MaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Masked language modeling (MLM) loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`) -- Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [DebertaForMaskedLM](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForMaskedLM) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, DebertaForMaskedLM >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-base") >>> model = DebertaForMaskedLM.from_pretrained("microsoft/deberta-base") >>> inputs = tokenizer("The capital of France is .", return_tensors="pt") >>> with torch.no_grad(): ... logits = model(**inputs).logits >>> # retrieve index of >>> mask_token_index = (inputs.input_ids == tokenizer.mask_token_id)[0].nonzero(as_tuple=True)[0] >>> predicted_token_id = logits[0, mask_token_index].argmax(axis=-1) >>> tokenizer.decode(predicted_token_id) ... >>> labels = tokenizer("The capital of France is Paris.", return_tensors="pt")["input_ids"] >>> # mask labels of non- tokens >>> labels = torch.where(inputs.input_ids == tokenizer.mask_token_id, labels, -100) >>> outputs = model(**inputs, labels=labels) >>> round(outputs.loss.item(), 2) ... ``` **Parameters:** config ([DebertaForMaskedLM](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForMaskedLM)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_outputs.MaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or `tuple(torch.FloatTensor)`` A [transformers.modeling_outputs.MaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Masked language modeling (MLM) loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`) -- Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## DebertaForSequenceClassification[[transformers.DebertaForSequenceClassification]][[transformers.DebertaForSequenceClassification]] #### transformers.DebertaForSequenceClassification[[transformers.DebertaForSequenceClassification]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L953) DeBERTa Model transformer with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks. This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.DebertaForSequenceClassification.forwardhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L978[{"name": "input_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "token_type_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "labels", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **labels** (`torch.LongTensor` of shape `(batch_size,)`, *optional*) -- Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If `config.num_labels > 1` a classification loss is computed (Cross-Entropy). - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[transformers.modeling_outputs.SequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.SequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [DebertaForSequenceClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForSequenceClassification) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example of single-label classification: ```python >>> import torch >>> from transformers import AutoTokenizer, DebertaForSequenceClassification >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-base") >>> model = DebertaForSequenceClassification.from_pretrained("microsoft/deberta-base") >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt") >>> with torch.no_grad(): ... logits = model(**inputs).logits >>> predicted_class_id = logits.argmax().item() >>> model.config.id2label[predicted_class_id] ... >>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)` >>> num_labels = len(model.config.id2label) >>> model = DebertaForSequenceClassification.from_pretrained("microsoft/deberta-base", num_labels=num_labels) >>> labels = torch.tensor([1]) >>> loss = model(**inputs, labels=labels).loss >>> round(loss.item(), 2) ... ``` Example of multi-label classification: ```python >>> import torch >>> from transformers import AutoTokenizer, DebertaForSequenceClassification >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-base") >>> model = DebertaForSequenceClassification.from_pretrained("microsoft/deberta-base", problem_type="multi_label_classification") >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt") >>> with torch.no_grad(): ... logits = model(**inputs).logits >>> predicted_class_ids = torch.arange(0, logits.shape[-1])[torch.sigmoid(logits).squeeze(dim=0) > 0.5] >>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)` >>> num_labels = len(model.config.id2label) >>> model = DebertaForSequenceClassification.from_pretrained( ... "microsoft/deberta-base", num_labels=num_labels, problem_type="multi_label_classification" ... ) >>> labels = torch.sum( ... torch.nn.functional.one_hot(predicted_class_ids[None, :].clone(), num_classes=num_labels), dim=1 ... ).to(torch.float) >>> loss = model(**inputs, labels=labels).loss ``` **Parameters:** config ([DebertaForSequenceClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForSequenceClassification)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_outputs.SequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`` A [transformers.modeling_outputs.SequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## DebertaForTokenClassification[[transformers.DebertaForTokenClassification]][[transformers.DebertaForTokenClassification]] #### transformers.DebertaForTokenClassification[[transformers.DebertaForTokenClassification]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L1060) The Deberta transformer with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.DebertaForTokenClassification.forwardhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L1072[{"name": "input_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "token_type_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "labels", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **labels** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) -- Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[transformers.modeling_outputs.TokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.TokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.num_labels)`) -- Classification scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [DebertaForTokenClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForTokenClassification) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, DebertaForTokenClassification >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-base") >>> model = DebertaForTokenClassification.from_pretrained("microsoft/deberta-base") >>> inputs = tokenizer( ... "HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="pt" ... ) >>> with torch.no_grad(): ... logits = model(**inputs).logits >>> predicted_token_class_ids = logits.argmax(-1) >>> # Note that tokens are classified rather then input words which means that >>> # there might be more predicted token classes than words. >>> # Multiple token classes might account for the same word >>> predicted_tokens_classes = [model.config.id2label[t.item()] for t in predicted_token_class_ids[0]] >>> predicted_tokens_classes ... >>> labels = predicted_token_class_ids >>> loss = model(**inputs, labels=labels).loss >>> round(loss.item(), 2) ... ``` **Parameters:** config ([DebertaForTokenClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForTokenClassification)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_outputs.TokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`` A [transformers.modeling_outputs.TokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification loss. - **logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.num_labels)`) -- Classification scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## DebertaForQuestionAnswering[[transformers.DebertaForQuestionAnswering]][[transformers.DebertaForQuestionAnswering]] #### transformers.DebertaForQuestionAnswering[[transformers.DebertaForQuestionAnswering]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L1122) The Deberta transformer with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layer on top of the hidden-states output to compute `span start logits` and `span end logits`). This model inherits from [PreTrainedModel](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.) This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior. forwardtransformers.DebertaForQuestionAnswering.forwardhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_deberta.py#L1133[{"name": "input_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "token_type_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "start_positions", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "end_positions", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`torch.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert `input_ids` indices into associated vectors than the model's internal embedding lookup matrix. - **start_positions** (`torch.Tensor` of shape `(batch_size,)`, *optional*) -- Labels for position (index) of the start of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence are not taken into account for computing the loss. - **end_positions** (`torch.Tensor` of shape `(batch_size,)`, *optional*) -- Labels for position (index) of the end of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence are not taken into account for computing the loss. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.0[transformers.modeling_outputs.QuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.QuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. - **start_logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`) -- Span-start scores (before SoftMax). - **end_logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`) -- Span-end scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [DebertaForQuestionAnswering](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForQuestionAnswering) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, DebertaForQuestionAnswering >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-base") >>> model = DebertaForQuestionAnswering.from_pretrained("microsoft/deberta-base") >>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet" >>> inputs = tokenizer(question, text, return_tensors="pt") >>> with torch.no_grad(): ... outputs = model(**inputs) >>> answer_start_index = outputs.start_logits.argmax() >>> answer_end_index = outputs.end_logits.argmax() >>> predict_answer_tokens = inputs.input_ids[0, answer_start_index : answer_end_index + 1] >>> tokenizer.decode(predict_answer_tokens, skip_special_tokens=True) ... >>> # target is "nice puppet" >>> target_start_index = torch.tensor([14]) >>> target_end_index = torch.tensor([15]) >>> outputs = model(**inputs, start_positions=target_start_index, end_positions=target_end_index) >>> loss = outputs.loss >>> round(loss.item(), 2) ... ``` **Parameters:** config ([DebertaForQuestionAnswering](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaForQuestionAnswering)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_outputs.QuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`` A [transformers.modeling_outputs.QuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or a tuple of `torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. - **start_logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`) -- Span-start scores (before SoftMax). - **end_logits** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`) -- Span-end scores (before SoftMax). - **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the optional initial embedding outputs. - **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## TFDebertaModel[[transformers.TFDebertaModel]][[transformers.TFDebertaModel]] #### transformers.TFDebertaModel[[transformers.TFDebertaModel]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1237) The bare DeBERTa Model transformer outputting raw hidden-states without any specific head on top. The DeBERTa model was proposed in [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://huggingface.co/papers/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen. It's build on top of BERT/RoBERTa with two improvements, i.e. disentangled attention and enhanced mask decoder. With those two improvements, it out perform BERT/RoBERTa on a majority of tasks with 80GB pretraining data. This model is also a [keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior. TensorFlow models and layers in `transformers` accept two formats as input: - having all inputs as keyword arguments (like PyTorch models), or - having all inputs as a list, tuple or dict in the first positional argument. The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first positional argument: - a single Tensor with `input_ids` only and nothing else: `model(input_ids)` - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])` - a dictionary with one or several input Tensors associated to the input names given in the docstring: `model({"input_ids": input_ids, "token_type_ids": token_type_ids})` Note that when creating models and layers with [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry about any of this, as you can just pass inputs like you would to any other Python function! calltransformers.TFDebertaModel.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1243[{"name": "input_ids", "val": ": TFModelInputType | None = None"}, {"name": "attention_mask", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "token_type_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "position_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "inputs_embeds", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "training", "val": ": bool | None = False"}]- **input_ids** (`np.ndarray`, `tf.Tensor`, `list[tf.Tensor]` ``dict[str, tf.Tensor]` or `dict[str, np.ndarray]` and each example must have the shape `(batch_size, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [`~utils.ModelOutput``] instead of a plain tuple.0[transformers.modeling_tf_outputs.TFBaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutput) or `tuple(tf.Tensor)`A [transformers.modeling_tf_outputs.TFBaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **last_hidden_state** (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model. - **hidden_states** (`tuple(tf.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [TFDebertaModel](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaModel) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, TFDebertaModel >>> import tensorflow as tf >>> tokenizer = AutoTokenizer.from_pretrained("kamalkraj/deberta-base") >>> model = TFDebertaModel.from_pretrained("kamalkraj/deberta-base") >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf") >>> outputs = model(inputs) >>> last_hidden_states = outputs.last_hidden_state ``` **Parameters:** config ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_tf_outputs.TFBaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutput) or `tuple(tf.Tensor)`` A [transformers.modeling_tf_outputs.TFBaseModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **last_hidden_state** (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model. - **hidden_states** (`tuple(tf.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## TFDebertaPreTrainedModel[[transformers.TFDebertaPreTrainedModel]][[transformers.TFDebertaPreTrainedModel]] #### transformers.TFDebertaPreTrainedModel[[transformers.TFDebertaPreTrainedModel]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1137) An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained models. calltransformers.TFDebertaPreTrainedModel.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/tf_keras/src/engine/training.py#L590[{"name": "inputs", "val": ""}, {"name": "training", "val": " = None"}, {"name": "mask", "val": " = None"}]- **inputs** -- Input tensor, or dict/list/tuple of input tensors. - **training** -- Boolean or boolean scalar tensor, indicating whether to run the `Network` in training mode or inference mode. - **mask** -- A mask or list of masks. A mask can be either a boolean tensor or None (no mask). For more details, check the guide [here](https://www.tensorflow.org/guide/tf_keras/masking_and_padding).0A tensor if there is a single output, or a list of tensors if there are more than one outputs. Calls the model on new inputs and returns the outputs as tensors. In this case `call()` just reapplies all ops in the graph to the new inputs (e.g. build a new computational graph from the provided inputs). Note: This method should not be called directly. It is only meant to be overridden when subclassing `tf.keras.Model`. To call a model on an input, always use the `__call__()` method, i.e. `model(inputs)`, which relies on the underlying `call()` method. **Parameters:** inputs : Input tensor, or dict/list/tuple of input tensors. training : Boolean or boolean scalar tensor, indicating whether to run the `Network` in training mode or inference mode. mask : A mask or list of masks. A mask can be either a boolean tensor or None (no mask). For more details, check the guide [here](https://www.tensorflow.org/guide/tf_keras/masking_and_padding). **Returns:** A tensor if there is a single output, or a list of tensors if there are more than one outputs. ## TFDebertaForMaskedLM[[transformers.TFDebertaForMaskedLM]][[transformers.TFDebertaForMaskedLM]] #### transformers.TFDebertaForMaskedLM[[transformers.TFDebertaForMaskedLM]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1286) DeBERTa Model with a `language modeling` head on top. The DeBERTa model was proposed in [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://huggingface.co/papers/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen. It's build on top of BERT/RoBERTa with two improvements, i.e. disentangled attention and enhanced mask decoder. With those two improvements, it out perform BERT/RoBERTa on a majority of tasks with 80GB pretraining data. This model is also a [keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior. TensorFlow models and layers in `transformers` accept two formats as input: - having all inputs as keyword arguments (like PyTorch models), or - having all inputs as a list, tuple or dict in the first positional argument. The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first positional argument: - a single Tensor with `input_ids` only and nothing else: `model(input_ids)` - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])` - a dictionary with one or several input Tensors associated to the input names given in the docstring: `model({"input_ids": input_ids, "token_type_ids": token_type_ids})` Note that when creating models and layers with [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry about any of this, as you can just pass inputs like you would to any other Python function! calltransformers.TFDebertaForMaskedLM.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1302[{"name": "input_ids", "val": ": TFModelInputType | None = None"}, {"name": "attention_mask", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "token_type_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "position_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "inputs_embeds", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "labels", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "training", "val": ": bool | None = False"}]- **input_ids** (`np.ndarray`, `tf.Tensor`, `list[tf.Tensor]` ``dict[str, tf.Tensor]` or `dict[str, np.ndarray]` and each example must have the shape `(batch_size, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [`~utils.ModelOutput``] instead of a plain tuple. - **labels** (`tf.Tensor` or `np.ndarray` of shape `(batch_size, sequence_length)`, *optional*) -- Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ..., config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`0[transformers.modeling_tf_outputs.TFMaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or `tuple(tf.Tensor)`A [transformers.modeling_tf_outputs.TFMaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(n,)`, *optional*, where n is the number of non-masked labels, returned when `labels` is provided) -- Masked language modeling (MLM) loss. - **logits** (`tf.Tensor` of shape `(batch_size, sequence_length, config.vocab_size)`) -- Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [TFDebertaForMaskedLM](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForMaskedLM) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, TFDebertaForMaskedLM >>> import tensorflow as tf >>> tokenizer = AutoTokenizer.from_pretrained("kamalkraj/deberta-base") >>> model = TFDebertaForMaskedLM.from_pretrained("kamalkraj/deberta-base") >>> inputs = tokenizer("The capital of France is [MASK].", return_tensors="tf") >>> logits = model(**inputs).logits >>> # retrieve index of [MASK] >>> mask_token_index = tf.where((inputs.input_ids == tokenizer.mask_token_id)[0]) >>> selected_logits = tf.gather_nd(logits[0], indices=mask_token_index) >>> predicted_token_id = tf.math.argmax(selected_logits, axis=-1) ``` ```python >>> labels = tokenizer("The capital of France is Paris.", return_tensors="tf")["input_ids"] >>> # mask labels of non-[MASK] tokens >>> labels = tf.where(inputs.input_ids == tokenizer.mask_token_id, labels, -100) >>> outputs = model(**inputs, labels=labels) ``` **Parameters:** config ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_tf_outputs.TFMaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or `tuple(tf.Tensor)`` A [transformers.modeling_tf_outputs.TFMaskedLMOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(n,)`, *optional*, where n is the number of non-masked labels, returned when `labels` is provided) -- Masked language modeling (MLM) loss. - **logits** (`tf.Tensor` of shape `(batch_size, sequence_length, config.vocab_size)`) -- Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## TFDebertaForSequenceClassification[[transformers.TFDebertaForSequenceClassification]][[transformers.TFDebertaForSequenceClassification]] #### transformers.TFDebertaForSequenceClassification[[transformers.TFDebertaForSequenceClassification]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1373) DeBERTa Model transformer with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks. The DeBERTa model was proposed in [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://huggingface.co/papers/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen. It's build on top of BERT/RoBERTa with two improvements, i.e. disentangled attention and enhanced mask decoder. With those two improvements, it out perform BERT/RoBERTa on a majority of tasks with 80GB pretraining data. This model is also a [keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior. TensorFlow models and layers in `transformers` accept two formats as input: - having all inputs as keyword arguments (like PyTorch models), or - having all inputs as a list, tuple or dict in the first positional argument. The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first positional argument: - a single Tensor with `input_ids` only and nothing else: `model(input_ids)` - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])` - a dictionary with one or several input Tensors associated to the input names given in the docstring: `model({"input_ids": input_ids, "token_type_ids": token_type_ids})` Note that when creating models and layers with [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry about any of this, as you can just pass inputs like you would to any other Python function! calltransformers.TFDebertaForSequenceClassification.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1392[{"name": "input_ids", "val": ": TFModelInputType | None = None"}, {"name": "attention_mask", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "token_type_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "position_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "inputs_embeds", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "labels", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "training", "val": ": bool | None = False"}]- **input_ids** (`np.ndarray`, `tf.Tensor`, `list[tf.Tensor]` ``dict[str, tf.Tensor]` or `dict[str, np.ndarray]` and each example must have the shape `(batch_size, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [`~utils.ModelOutput``] instead of a plain tuple. - **labels** (`tf.Tensor` or `np.ndarray` of shape `(batch_size,)`, *optional*) -- Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If `config.num_labels > 1` a classification loss is computed (Cross-Entropy).0[transformers.modeling_tf_outputs.TFSequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or `tuple(tf.Tensor)`A [transformers.modeling_tf_outputs.TFSequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(batch_size, )`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss. - **logits** (`tf.Tensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [TFDebertaForSequenceClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForSequenceClassification) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, TFDebertaForSequenceClassification >>> import tensorflow as tf >>> tokenizer = AutoTokenizer.from_pretrained("kamalkraj/deberta-base") >>> model = TFDebertaForSequenceClassification.from_pretrained("kamalkraj/deberta-base") >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf") >>> logits = model(**inputs).logits >>> predicted_class_id = int(tf.math.argmax(logits, axis=-1)[0]) ``` ```python >>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)` >>> num_labels = len(model.config.id2label) >>> model = TFDebertaForSequenceClassification.from_pretrained("kamalkraj/deberta-base", num_labels=num_labels) >>> labels = tf.constant(1) >>> loss = model(**inputs, labels=labels).loss ``` **Parameters:** config ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_tf_outputs.TFSequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or `tuple(tf.Tensor)`` A [transformers.modeling_tf_outputs.TFSequenceClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(batch_size, )`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss. - **logits** (`tf.Tensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## TFDebertaForTokenClassification[[transformers.TFDebertaForTokenClassification]][[transformers.TFDebertaForTokenClassification]] #### transformers.TFDebertaForTokenClassification[[transformers.TFDebertaForTokenClassification]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1472) DeBERTa Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. The DeBERTa model was proposed in [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://huggingface.co/papers/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen. It's build on top of BERT/RoBERTa with two improvements, i.e. disentangled attention and enhanced mask decoder. With those two improvements, it out perform BERT/RoBERTa on a majority of tasks with 80GB pretraining data. This model is also a [keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior. TensorFlow models and layers in `transformers` accept two formats as input: - having all inputs as keyword arguments (like PyTorch models), or - having all inputs as a list, tuple or dict in the first positional argument. The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first positional argument: - a single Tensor with `input_ids` only and nothing else: `model(input_ids)` - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])` - a dictionary with one or several input Tensors associated to the input names given in the docstring: `model({"input_ids": input_ids, "token_type_ids": token_type_ids})` Note that when creating models and layers with [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry about any of this, as you can just pass inputs like you would to any other Python function! calltransformers.TFDebertaForTokenClassification.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1485[{"name": "input_ids", "val": ": TFModelInputType | None = None"}, {"name": "attention_mask", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "token_type_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "position_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "inputs_embeds", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "labels", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "training", "val": ": bool | None = False"}]- **input_ids** (`np.ndarray`, `tf.Tensor`, `list[tf.Tensor]` ``dict[str, tf.Tensor]` or `dict[str, np.ndarray]` and each example must have the shape `(batch_size, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [`~utils.ModelOutput``] instead of a plain tuple. - **labels** (`tf.Tensor` or `np.ndarray` of shape `(batch_size, sequence_length)`, *optional*) -- Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.0[transformers.modeling_tf_outputs.TFTokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or `tuple(tf.Tensor)`A [transformers.modeling_tf_outputs.TFTokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(n,)`, *optional*, where n is the number of unmasked labels, returned when `labels` is provided) -- Classification loss. - **logits** (`tf.Tensor` of shape `(batch_size, sequence_length, config.num_labels)`) -- Classification scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [TFDebertaForTokenClassification](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForTokenClassification) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, TFDebertaForTokenClassification >>> import tensorflow as tf >>> tokenizer = AutoTokenizer.from_pretrained("kamalkraj/deberta-base") >>> model = TFDebertaForTokenClassification.from_pretrained("kamalkraj/deberta-base") >>> inputs = tokenizer( ... "HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="tf" ... ) >>> logits = model(**inputs).logits >>> predicted_token_class_ids = tf.math.argmax(logits, axis=-1) >>> # Note that tokens are classified rather then input words which means that >>> # there might be more predicted token classes than words. >>> # Multiple token classes might account for the same word >>> predicted_tokens_classes = [model.config.id2label[t] for t in predicted_token_class_ids[0].numpy().tolist()] ``` ```python >>> labels = predicted_token_class_ids >>> loss = tf.math.reduce_mean(model(**inputs, labels=labels).loss) ``` **Parameters:** config ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_tf_outputs.TFTokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or `tuple(tf.Tensor)`` A [transformers.modeling_tf_outputs.TFTokenClassifierOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(n,)`, *optional*, where n is the number of unmasked labels, returned when `labels` is provided) -- Classification loss. - **logits** (`tf.Tensor` of shape `(batch_size, sequence_length, config.num_labels)`) -- Classification scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. ## TFDebertaForQuestionAnswering[[transformers.TFDebertaForQuestionAnswering]][[transformers.TFDebertaForQuestionAnswering]] #### transformers.TFDebertaForQuestionAnswering[[transformers.TFDebertaForQuestionAnswering]] [Source](https://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1555) DeBERTa Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of the hidden-states output to compute `span start logits` and `span end logits`). The DeBERTa model was proposed in [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://huggingface.co/papers/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen. It's build on top of BERT/RoBERTa with two improvements, i.e. disentangled attention and enhanced mask decoder. With those two improvements, it out perform BERT/RoBERTa on a majority of tasks with 80GB pretraining data. This model is also a [keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior. TensorFlow models and layers in `transformers` accept two formats as input: - having all inputs as keyword arguments (like PyTorch models), or - having all inputs as a list, tuple or dict in the first positional argument. The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first positional argument: - a single Tensor with `input_ids` only and nothing else: `model(input_ids)` - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])` - a dictionary with one or several input Tensors associated to the input names given in the docstring: `model({"input_ids": input_ids, "token_type_ids": token_type_ids})` Note that when creating models and layers with [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry about any of this, as you can just pass inputs like you would to any other Python function! calltransformers.TFDebertaForQuestionAnswering.callhttps://github.com/huggingface/transformers/blob/v4.57.1/src/transformers/models/deberta/modeling_tf_deberta.py#L1567[{"name": "input_ids", "val": ": TFModelInputType | None = None"}, {"name": "attention_mask", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "token_type_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "position_ids", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "inputs_embeds", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "output_attentions", "val": ": bool | None = None"}, {"name": "output_hidden_states", "val": ": bool | None = None"}, {"name": "return_dict", "val": ": bool | None = None"}, {"name": "start_positions", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "end_positions", "val": ": np.ndarray | tf.Tensor | None = None"}, {"name": "training", "val": ": bool | None = False"}]- **input_ids** (`np.ndarray`, `tf.Tensor`, `list[tf.Tensor]` ``dict[str, tf.Tensor]` or `dict[str, np.ndarray]` and each example must have the shape `(batch_size, sequence_length)`) -- Indices of input sequence tokens in the vocabulary. Indices can be obtained using [AutoTokenizer](/docs/transformers/v4.57.1/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v4.57.1/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#input-ids) - **attention_mask** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`: - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**. [What are attention masks?](../glossary#attention-mask) - **token_type_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0, 1]`: - 0 corresponds to a *sentence A* token, - 1 corresponds to a *sentence B* token. [What are token type IDs?](../glossary#token-type-ids) - **position_ids** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.max_position_embeddings - 1]`. [What are position IDs?](../glossary#position-ids) - **inputs_embeds** (`np.ndarray` or `tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) -- Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert *input_ids* indices into associated vectors than the model's internal embedding lookup matrix. - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [`~utils.ModelOutput``] instead of a plain tuple. - **start_positions** (`tf.Tensor` or `np.ndarray` of shape `(batch_size,)`, *optional*) -- Labels for position (index) of the start of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence are not taken into account for computing the loss. - **end_positions** (`tf.Tensor` or `np.ndarray` of shape `(batch_size,)`, *optional*) -- Labels for position (index) of the end of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence are not taken into account for computing the loss.0[transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or `tuple(tf.Tensor)`A [transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(batch_size, )`, *optional*, returned when `start_positions` and `end_positions` are provided) -- Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. - **start_logits** (`tf.Tensor` of shape `(batch_size, sequence_length)`) -- Span-start scores (before SoftMax). - **end_logits** (`tf.Tensor` of shape `(batch_size, sequence_length)`) -- Span-end scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads. The [TFDebertaForQuestionAnswering](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.TFDebertaForQuestionAnswering) forward method, overrides the `__call__` special method. Although the recipe for forward pass needs to be defined within this function, one should call the `Module` instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them. Example: ```python >>> from transformers import AutoTokenizer, TFDebertaForQuestionAnswering >>> import tensorflow as tf >>> tokenizer = AutoTokenizer.from_pretrained("kamalkraj/deberta-base") >>> model = TFDebertaForQuestionAnswering.from_pretrained("kamalkraj/deberta-base") >>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet" >>> inputs = tokenizer(question, text, return_tensors="tf") >>> outputs = model(**inputs) >>> answer_start_index = int(tf.math.argmax(outputs.start_logits, axis=-1)[0]) >>> answer_end_index = int(tf.math.argmax(outputs.end_logits, axis=-1)[0]) >>> predict_answer_tokens = inputs.input_ids[0, answer_start_index : answer_end_index + 1] ``` ```python >>> # target is "nice puppet" >>> target_start_index = tf.constant([14]) >>> target_end_index = tf.constant([15]) >>> outputs = model(**inputs, start_positions=target_start_index, end_positions=target_end_index) >>> loss = tf.math.reduce_mean(outputs.loss) ``` **Parameters:** config ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v4.57.1/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** `[transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or `tuple(tf.Tensor)`` A [transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput](/docs/transformers/v4.57.1/ko/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or a tuple of `tf.Tensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various elements depending on the configuration ([DebertaConfig](/docs/transformers/v4.57.1/ko/model_doc/deberta#transformers.DebertaConfig)) and inputs. - **loss** (`tf.Tensor` of shape `(batch_size, )`, *optional*, returned when `start_positions` and `end_positions` are provided) -- Total span extraction loss is the sum of a Cross-Entropy for the start and end positions. - **start_logits** (`tf.Tensor` of shape `(batch_size, sequence_length)`) -- Span-start scores (before SoftMax). - **end_logits** (`tf.Tensor` of shape `(batch_size, sequence_length)`) -- Span-end scores (before SoftMax). - **hidden_states** (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `tf.Tensor` (one for the output of the embeddings + one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states of the model at the output of each layer plus the initial embedding outputs. - **attentions** (`tuple(tf.Tensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `tf.Tensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`. Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.