# RoBERTa

## RoBERTa

[![Models](https://img.shields.io/badge/All_model_pages-roberta-blueviolet)](https://huggingface.co/models?filter=roberta)[![Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/docs-demos/roberta-base)[![Paper page](https://img.shields.io/badge/Paper%20page-1907.11692-green)](https://huggingface.co/papers/1907.11692)

### Overview

The RoBERTa model was proposed in [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, [Myle Ott](https://huggingface.co/myleott), Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov. It is based on Google’s BERT model released in 2018.

It builds on BERT and modifies key hyperparameters, removing the next-sentence pretraining objective and training with much larger mini-batches and learning rates.

The abstract from the paper is the following:

*Language model pretraining has led to significant performance gains but careful comparison between different approaches is challenging. Training is computationally expensive, often done on private datasets of different sizes, and, as we will show, hyperparameter choices have significant impact on the final results. We present a replication study of BERT pretraining (Devlin et al., 2019) that carefully measures the impact of many key hyperparameters and training data size. We find that BERT was significantly undertrained, and can match or exceed the performance of every model published after it. Our best model achieves state-of-the-art results on GLUE, RACE and SQuAD. These results highlight the importance of previously overlooked design choices, and raise questions about the source of recently reported improvements. We release our models and code.*

Tips:

* This implementation is the same as [BertModel](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bert#transformers.BertModel) with a tiny embeddings tweak as well as a setup for Roberta pretrained models.
* RoBERTa has the same architecture as BERT, but uses a byte-level BPE as a tokenizer (same as GPT-2) and uses a different pretraining scheme.
* RoBERTa doesn’t have `token_type_ids`, you don’t need to indicate which token belongs to which segment. Just separate your segments with the separation token `tokenizer.sep_token` (or `</s>`)
* Same as BERT with better pretraining tricks:
  * dynamic masking: tokens are masked differently at each epoch, whereas BERT does it once and for all
  * together to reach 512 tokens (so the sentences are in an order than may span several documents)
  * train with larger batches
  * use BPE with bytes as a subunit and not characters (because of unicode characters)
* [CamemBERT](https://huggingface.co/docs/transformers/model_doc/camembert) is a wrapper around RoBERTa. Refer to this page for usage examples.

This model was contributed by [julien-c](https://huggingface.co/julien-c). The original code can be found [here](https://github.com/pytorch/fairseq/tree/master/examples/roberta).

### Resources

A list of official BOINC AI and community (indicated by 🌎) resources to help you get started with RoBERTa. If you’re interested in submitting a resource to be included here, please feel free to open a Pull Request and we’ll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.

Text Classification

* A blog on [Getting Started with Sentiment Analysis on Twitter](https://huggingface.co/blog/sentiment-analysis-twitter) using RoBERTa and the [Inference API](https://huggingface.co/inference-api).
* A blog on [Opinion Classification with Kili and BOINC AI AutoTrain](https://huggingface.co/blog/opinion-classification-with-kili) using RoBERTa.
* A notebook on how to [finetune RoBERTa for sentiment analysis](https://colab.research.google.com/github/DhavalTaunk08/NLP_scripts/blob/master/sentiment_analysis_using_roberta.ipynb). 🌎
* [RobertaForSequenceClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForSequenceClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb).
* [TFRobertaForSequenceClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForSequenceClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb).
* [FlaxRobertaForSequenceClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.FlaxRobertaForSequenceClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification_flax.ipynb).
* [Text classification task guide](https://huggingface.co/docs/transformers/tasks/sequence_classification)

Token Classification

* [RobertaForTokenClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForTokenClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/token-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb).
* [TFRobertaForTokenClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForTokenClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/token-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb).
* [FlaxRobertaForTokenClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.FlaxRobertaForTokenClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/token-classification).
* [Token classification](https://huggingface.co/course/chapter7/2?fw=pt) chapter of the 🌎 BOINC AI Course.
* [Token classification task guide](https://huggingface.co/docs/transformers/tasks/token_classification)

Fill-Mask

* A blog on [How to train a new language model from scratch using Transformers and Tokenizers](https://huggingface.co/blog/how-to-train) with RoBERTa.
* [RobertaForMaskedLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForMaskedLM) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/language-modeling#robertabertdistilbert-and-masked-language-modeling) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling.ipynb).
* [TFRobertaForMaskedLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForMaskedLM) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/language-modeling#run_mlmpy) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb).
* [FlaxRobertaForMaskedLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.FlaxRobertaForMaskedLM) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/language-modeling#masked-language-modeling) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/masked_language_modeling_flax.ipynb).
* [Masked language modeling](https://huggingface.co/course/chapter7/3?fw=pt) chapter of the 🌎 BOINC AI Course.
* [Masked language modeling task guide](https://huggingface.co/docs/transformers/tasks/masked_language_modeling)

Question Answering

* A blog on [Accelerated Inference with Optimum and Transformers Pipelines](https://huggingface.co/blog/optimum-inference) with RoBERTa for question answering.
* [RobertaForQuestionAnswering](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForQuestionAnswering) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb).
* [TFRobertaForQuestionAnswering](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForQuestionAnswering) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/question-answering) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb).
* [FlaxRobertaForQuestionAnswering](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.FlaxRobertaForQuestionAnswering) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/question-answering).
* [Question answering](https://huggingface.co/course/chapter7/7?fw=pt) chapter of the 🌎 BOINC AI Course.
* [Question answering task guide](https://huggingface.co/docs/transformers/tasks/question_answering)

**Multiple choice**

* [RobertaForMultipleChoice](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForMultipleChoice) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/multiple-choice) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice.ipynb).
* [TFRobertaForMultipleChoice](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForMultipleChoice) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/multiple-choice) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice-tf.ipynb).
* [Multiple choice task guide](https://huggingface.co/docs/transformers/tasks/multiple_choice)

### RobertaConfig

#### class transformers.RobertaForCausalLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L875)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa Model with a `language modeling` head on top for CLM fine-tuning.

This model inherits from [PreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L896)

( input\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Noneencoder\_hidden\_states: typing.Optional\[torch.FloatTensor] = Noneencoder\_attention\_mask: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Nonepast\_key\_values: typing.Tuple\[typing.Tuple\[torch.FloatTensor]] = Noneuse\_cache: typing.Optional\[bool] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.CausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.CausalLMOutputWithCrossAttentions) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **encoder\_hidden\_states** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) — Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder.
* **encoder\_attention\_mask** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) — Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
  * 1 for tokens that are **not masked**,
  * 0 for tokens that are **masked**.
* **labels** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) — Labels for computing the left-to-right language modeling loss (next word prediction). 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]`
* **past\_key\_values** (`tuple(tuple(torch.FloatTensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`) — Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.

  If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that don’t have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `decoder_input_ids` of shape `(batch_size, sequence_length)`.
* **use\_cache** (`bool`, *optional*) — If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see `past_key_values`).

Returns

[transformers.modeling\_outputs.CausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.CausalLMOutputWithCrossAttentions) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.CausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.CausalLMOutputWithCrossAttentions) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) — Language modeling loss (for next-token prediction).
* **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.
* **cross\_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)`.

  Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
* **past\_key\_values** (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`) — Tuple of `torch.FloatTensor` tuples of length `config.n_layers`, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant if `config.is_decoder = True`.

  Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.

The [RobertaForCausalLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForCausalLM) 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:

Copied

```
>>> from transformers import AutoTokenizer, RobertaForCausalLM, AutoConfig
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> config = AutoConfig.from_pretrained("roberta-base")
>>> config.is_decoder = True
>>> model = RobertaForCausalLM.from_pretrained("roberta-base", config=config)

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)

>>> prediction_logits = outputs.logits
```

### RobertaForMaskedLM

#### class transformers.RobertaForMaskedLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1026)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa Model with a `language modeling` head on top.

This model inherits from [PreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1050)

( input\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Noneencoder\_hidden\_states: typing.Optional\[torch.FloatTensor] = Noneencoder\_attention\_mask: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.MaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **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]`
* **kwargs** (`Dict[str, any]`, optional, defaults to *{}*) — Used to hide legacy arguments that have been deprecated.

Returns

[transformers.modeling\_outputs.MaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.MaskedLMOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.MaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [RobertaForMaskedLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForMaskedLM) 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:

Copied

```
>>> from transformers import AutoTokenizer, RobertaForMaskedLM
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = RobertaForMaskedLM.from_pretrained("roberta-base")

>>> inputs = tokenizer("The capital of France is <mask>.", return_tensors="pt")

>>> with torch.no_grad():
...     logits = model(**inputs).logits

>>> # retrieve index of <mask>
>>> 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)
' Paris'

>>> labels = tokenizer("The capital of France is Paris.", return_tensors="pt")["input_ids"]
>>> # mask labels of non-<mask> tokens
>>> labels = torch.where(inputs.input_ids == tokenizer.mask_token_id, labels, -100)

>>> outputs = model(**inputs, labels=labels)
>>> round(outputs.loss.item(), 2)
0.1
```

### RobertaForSequenceClassification

#### class transformers.RobertaForSequenceClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1157)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa 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](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1169)

( input\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.SequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **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).

Returns

[transformers.modeling\_outputs.SequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.SequenceClassifierOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.SequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [RobertaForSequenceClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForSequenceClassification) 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:

Copied

```
>>> import torch
>>> from transformers import AutoTokenizer, RobertaForSequenceClassification

>>> tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-emotion")
>>> model = RobertaForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-emotion")

>>> 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]
'optimism'

>>> # 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 = RobertaForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-emotion", num_labels=num_labels)

>>> labels = torch.tensor([1])
>>> loss = model(**inputs, labels=labels).loss
>>> round(loss.item(), 2)
0.08
```

Example of multi-label classification:

Copied

```
>>> import torch
>>> from transformers import AutoTokenizer, RobertaForSequenceClassification

>>> tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-emotion")
>>> model = RobertaForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-emotion", 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 = RobertaForSequenceClassification.from_pretrained(
...     "cardiffnlp/twitter-roberta-base-emotion", 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
```

### RobertaForMultipleChoice

#### class transformers.RobertaForMultipleChoice

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1256)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

Roberta Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a softmax) e.g. for RocStories/SWAG tasks.

This model inherits from [PreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1267)

( input\_ids: typing.Optional\[torch.LongTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.MultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` of shape `(batch_size, num_choices, 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **labels** (`torch.LongTensor` of shape `(batch_size,)`, *optional*) — Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See `input_ids` above)

Returns

[transformers.modeling\_outputs.MultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.MultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.MultipleChoiceModelOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **loss** (`torch.FloatTensor` of shape *(1,)*, *optional*, returned when `labels` is provided) — Classification loss.
* **logits** (`torch.FloatTensor` of shape `(batch_size, num_choices)`) — *num\_choices* is the second dimension of the input tensors. (see *input\_ids* above).

  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 [RobertaForMultipleChoice](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForMultipleChoice) 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:

Copied

```
>>> from transformers import AutoTokenizer, RobertaForMultipleChoice
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = RobertaForMultipleChoice.from_pretrained("roberta-base")

>>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
>>> choice0 = "It is eaten with a fork and a knife."
>>> choice1 = "It is eaten while held in the hand."
>>> labels = torch.tensor(0).unsqueeze(0)  # choice0 is correct (according to Wikipedia ;)), batch size 1

>>> encoding = tokenizer([prompt, prompt], [choice0, choice1], return_tensors="pt", padding=True)
>>> outputs = model(**{k: v.unsqueeze(0) for k, v in encoding.items()}, labels=labels)  # batch size is 1

>>> # the linear classifier still needs to be trained
>>> loss = outputs.loss
>>> logits = outputs.logits
```

### RobertaForTokenClassification

#### class transformers.RobertaForTokenClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1348)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

Roberta 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.

This model inherits from [PreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1363)

( input\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.TokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **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]`.

Returns

[transformers.modeling\_outputs.TokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.TokenClassifierOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.TokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [RobertaForTokenClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForTokenClassification) 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:

Copied

```
>>> from transformers import AutoTokenizer, RobertaForTokenClassification
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("Jean-Baptiste/roberta-large-ner-english")
>>> model = RobertaForTokenClassification.from_pretrained("Jean-Baptiste/roberta-large-ner-english")

>>> inputs = tokenizer(
...     "BOINCAI 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
['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']

>>> labels = predicted_token_class_ids
>>> loss = model(**inputs, labels=labels).loss
>>> round(loss.item(), 2)
0.01
```

### RobertaForQuestionAnswering

#### class transformers.RobertaForQuestionAnswering

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1455)

( config )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

Roberta 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`).

This model inherits from [PreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_roberta.py#L1466)

( input\_ids: typing.Optional\[torch.LongTensor] = Noneattention\_mask: typing.Optional\[torch.FloatTensor] = Nonetoken\_type\_ids: typing.Optional\[torch.LongTensor] = Noneposition\_ids: typing.Optional\[torch.LongTensor] = Nonehead\_mask: typing.Optional\[torch.FloatTensor] = Noneinputs\_embeds: typing.Optional\[torch.FloatTensor] = Nonestart\_positions: typing.Optional\[torch.LongTensor] = Noneend\_positions: typing.Optional\[torch.LongTensor] = Noneoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.QuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`torch.FloatTensor` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`torch.LongTensor` 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. This parameter can only be used when the model is initialized with `type_vocab_size` parameter with value

    > \= 2. All the value in this tensor should be always < type\_vocab\_size.

  [What are token type IDs?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`torch.LongTensor` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`torch.FloatTensor` 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](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **start\_positions** (`torch.LongTensor` 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.LongTensor` 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.

Returns

[transformers.modeling\_outputs.QuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.QuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.QuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [RobertaForQuestionAnswering](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaForQuestionAnswering) 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:

Copied

```
>>> from transformers import AutoTokenizer, RobertaForQuestionAnswering
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")
>>> model = RobertaForQuestionAnswering.from_pretrained("deepset/roberta-base-squad2")

>>> 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)
' puppet'

>>> # 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)
0.86
```

### TFRobertaModel

#### class transformers.TFRobertaModel

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L881)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

The bare RoBERTa Model transformer outputting raw hidden-states without any specific head on top.

This model inherits from [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L886)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneencoder\_hidden\_states: np.ndarray | tf.Tensor | None = Noneencoder\_attention\_mask: np.ndarray | tf.Tensor | None = Nonepast\_key\_values: Optional\[Tuple\[Tuple\[Union\[np.ndarray, tf.Tensor]]]] = Noneuse\_cache: Optional\[bool] = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **encoder\_hidden\_states** (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) — Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder.
* **encoder\_attention\_mask** (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) — Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
  * 1 for tokens that are **not masked**,
  * 0 for tokens that are **masked**.
* **past\_key\_values** (`Tuple[Tuple[tf.Tensor]]` of length `config.n_layers`) — contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding. If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that don’t have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `decoder_input_ids` of shape `(batch_size, sequence_length)`.
* **use\_cache** (`bool`, *optional*, defaults to `True`) — If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see `past_key_values`). Set to `False` during training, `True` during generation

Returns

[transformers.modeling\_tf\_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndCrossAttentions) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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.
* **pooler\_output** (`tf.Tensor` of shape `(batch_size, hidden_size)`) — Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  This output is usually *not* a good summary of the semantic content of the input, you’re often better with averaging or pooling the sequence of hidden-states for the whole input sequence.
* **past\_key\_values** (`List[tf.Tensor]`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`) — List of `tf.Tensor` of length `config.n_layers`, with each tensor of shape `(2, batch_size, num_heads, sequence_length, embed_size_per_head)`).

  Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
* **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.
* **cross\_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 of the decoder’s cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.

The [TFRobertaModel](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaModel) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaModel
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = TFRobertaModel.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf")
>>> outputs = model(inputs)

>>> last_hidden_states = outputs.last_hidden_state
```

### TFRobertaForCausalLM

#### class transformers.TFRobertaForCausalLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1080)

( \*args\*\*kwargs )

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1113)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneencoder\_hidden\_states: np.ndarray | tf.Tensor | None = Noneencoder\_attention\_mask: np.ndarray | tf.Tensor | None = Nonepast\_key\_values: Optional\[Tuple\[Tuple\[Union\[np.ndarray, tf.Tensor]]]] = Noneuse\_cache: Optional\[bool] = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonelabels: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFCausalLMOutputWithCrossAttentions) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **encoder\_hidden\_states** (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) — Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder.
* **encoder\_attention\_mask** (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) — Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
  * 1 for tokens that are **not masked**,
  * 0 for tokens that are **masked**.
* **past\_key\_values** (`Tuple[Tuple[tf.Tensor]]` of length `config.n_layers`) — contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding. If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that don’t have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `decoder_input_ids` of shape `(batch_size, sequence_length)`.
* **use\_cache** (`bool`, *optional*, defaults to `True`) — If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see `past_key_values`). Set to `False` during training, `True` during generation
* **labels** (`tf.Tensor` or `np.ndarray` of shape `(batch_size, sequence_length)`, *optional*) — Labels for computing the cross entropy classification loss. Indices should be in `[0, ..., config.vocab_size - 1]`.

Returns

[transformers.modeling\_tf\_outputs.TFCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFCausalLMOutputWithCrossAttentions) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFCausalLMOutputWithCrossAttentions) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **loss** (`tf.Tensor` of shape `(n,)`, *optional*, where n is the number of non-masked labels, returned when `labels` is provided) — Language modeling loss (for next-token prediction).
* **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.
* **cross\_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 of the decoder’s cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
* **past\_key\_values** (`List[tf.Tensor]`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`) — List of `tf.Tensor` of length `config.n_layers`, with each tensor of shape `(2, batch_size, num_heads, sequence_length, embed_size_per_head)`).

  Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.

The [TFRobertaForCausalLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForCausalLM) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForCausalLM
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = TFRobertaForCausalLM.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf")
>>> outputs = model(inputs)
>>> logits = outputs.logits
```

### TFRobertaForMaskedLM

#### class transformers.TFRobertaForMaskedLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1003)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa Model with a `language modeling` head on top.

This model inherits from [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1020)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonelabels: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFMaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **labels** (`tf.Tensor` 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]`

Returns

[transformers.modeling\_tf\_outputs.TFMaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFMaskedLMOutput) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFMaskedLMOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [TFRobertaForMaskedLM](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForMaskedLM) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForMaskedLM
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = TFRobertaForMaskedLM.from_pretrained("roberta-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)
>>> tokenizer.decode(predicted_token_id)
' Paris'
```

Copied

```
>>> 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)
>>> round(float(outputs.loss), 2)
0.1
```

### TFRobertaForSequenceClassification

#### class transformers.TFRobertaForSequenceClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1237)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa 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 [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1248)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonelabels: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **labels** (`tf.Tensor` 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).

Returns

[transformers.modeling\_tf\_outputs.TFSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFSequenceClassifierOutput) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [TFRobertaForSequenceClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForSequenceClassification) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForSequenceClassification
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-emotion")
>>> model = TFRobertaForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-emotion")

>>> 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])
>>> model.config.id2label[predicted_class_id]
'optimism'
```

Copied

```
>>> # 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 = TFRobertaForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-emotion", num_labels=num_labels)

>>> labels = tf.constant(1)
>>> loss = model(**inputs, labels=labels).loss
>>> round(float(loss), 2)
0.08
```

### TFRobertaForMultipleChoice

#### class transformers.TFRobertaForMultipleChoice

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1313)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

Roberta Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a softmax) e.g. for RocStories/SWAG tasks.

This model inherits from [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1327)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonelabels: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFMultipleChoiceModelOutput) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, num_choices, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` or `tf.Tensor` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`tf.Tensor` of shape `(batch_size, num_choices, 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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **labels** (`tf.Tensor` of shape `(batch_size,)`, *optional*) — Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]` where `num_choices` is the size of the second dimension of the input tensors. (See `input_ids` above)

Returns

[transformers.modeling\_tf\_outputs.TFMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFMultipleChoiceModelOutput) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFMultipleChoiceModelOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **loss** (`tf.Tensor` of shape *(batch\_size, )*, *optional*, returned when `labels` is provided) — Classification loss.
* **logits** (`tf.Tensor` of shape `(batch_size, num_choices)`) — *num\_choices* is the second dimension of the input tensors. (see *input\_ids* above).

  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 [TFRobertaForMultipleChoice](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForMultipleChoice) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForMultipleChoice
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = TFRobertaForMultipleChoice.from_pretrained("roberta-base")

>>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
>>> choice0 = "It is eaten with a fork and a knife."
>>> choice1 = "It is eaten while held in the hand."

>>> encoding = tokenizer([prompt, prompt], [choice0, choice1], return_tensors="tf", padding=True)
>>> inputs = {k: tf.expand_dims(v, 0) for k, v in encoding.items()}
>>> outputs = model(inputs)  # batch size is 1

>>> # the linear classifier still needs to be trained
>>> logits = outputs.logits
```

### TFRobertaForTokenClassification

#### class transformers.TFRobertaForTokenClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1403)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa 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.

This model inherits from [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1421)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonelabels: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **labels** (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*) — Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.

Returns

[transformers.modeling\_tf\_outputs.TFTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFTokenClassifierOutput) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [TFRobertaForTokenClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForTokenClassification) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForTokenClassification
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("ydshieh/roberta-large-ner-english")
>>> model = TFRobertaForTokenClassification.from_pretrained("ydshieh/roberta-large-ner-english")

>>> inputs = tokenizer(
...     "BOINCAI 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()]
>>> predicted_tokens_classes
['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']
```

Copied

```
>>> labels = predicted_token_class_ids
>>> loss = tf.math.reduce_mean(model(**inputs, labels=labels).loss)
>>> round(float(loss), 2)
0.01
```

### TFRobertaForQuestionAnswering

#### class transformers.TFRobertaForQuestionAnswering

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1486)

( \*args\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa 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`).

This model inherits from [TFPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.TFPreTrainedModel). 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 [tf.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!

**call**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_tf_roberta.py#L1499)

( input\_ids: TFModelInputType | None = Noneattention\_mask: np.ndarray | tf.Tensor | None = Nonetoken\_type\_ids: np.ndarray | tf.Tensor | None = Noneposition\_ids: np.ndarray | tf.Tensor | None = Nonehead\_mask: np.ndarray | tf.Tensor | None = Noneinputs\_embeds: np.ndarray | tf.Tensor | None = Noneoutput\_attentions: Optional\[bool] = Noneoutput\_hidden\_states: Optional\[bool] = Nonereturn\_dict: Optional\[bool] = Nonestart\_positions: np.ndarray | tf.Tensor | None = Noneend\_positions: np.ndarray | tf.Tensor | None = Nonetraining: Optional\[bool] = False ) → [transformers.modeling\_tf\_outputs.TFQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or `tuple(tf.Tensor)`

Parameters

* **input\_ids** (`Numpy array` or `tf.Tensor` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) and [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`Numpy array` 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?](https://huggingface.co/docs/transformers/glossary#position-ids)
* **head\_mask** (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*) — Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **inputs\_embeds** (`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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **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. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
* **training** (`bool`, *optional*, defaults to `False`) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).
* **start\_positions** (`tf.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** (`tf.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.

Returns

[transformers.modeling\_tf\_outputs.TFQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_tf_outputs.TFQuestionAnsweringModelOutput) or `tuple(tf.Tensor)`

A [transformers.modeling\_tf\_outputs.TFQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) 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 [TFRobertaForQuestionAnswering](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.TFRobertaForQuestionAnswering) 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:

Copied

```
>>> from transformers import AutoTokenizer, TFRobertaForQuestionAnswering
>>> import tensorflow as tf

>>> tokenizer = AutoTokenizer.from_pretrained("ydshieh/roberta-base-squad2")
>>> model = TFRobertaForQuestionAnswering.from_pretrained("ydshieh/roberta-base-squad2")

>>> 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]
>>> tokenizer.decode(predict_answer_tokens)
' puppet'
```

Copied

```
>>> # 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)
>>> round(float(loss), 2)
0.86
```

### FlaxRobertaModel

#### class transformers.FlaxRobertaModel

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L990)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

The bare RoBERTa Model transformer outputting raw hidden-states without any specific head on top.

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **last\_hidden\_state** (`jnp.ndarray` of shape `(batch_size, sequence_length, hidden_size)`) — Sequence of hidden-states at the output of the last layer of the model.
* **pooler\_output** (`jnp.ndarray` of shape `(batch_size, hidden_size)`) — Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaModel

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaModel.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="jax")
>>> outputs = model(**inputs)

>>> last_hidden_states = outputs.last_hidden_state
```

### FlaxRobertaForCausalLM

#### class transformers.FlaxRobertaForCausalLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1452)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

Roberta Model with a language modeling head on top (a linear layer on top of the hidden-states output) e.g for autoregressive tasks.

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxCausalLMOutputWithCrossAttentions](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxCausalLMOutputWithCrossAttentions) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **logits** (`jnp.ndarray` 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(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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.
* **cross\_attentions** (`tuple(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`.

  Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
* **past\_key\_values** (`tuple(tuple(jnp.ndarray))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`) — Tuple of `jnp.ndarray` tuples of length `config.n_layers`, with each tuple containing the cached key, value states of the self-attention and the cross-attention layers if model is used in encoder-decoder setting. Only relevant if `config.is_decoder = True`.

  Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.

The `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForCausalLM

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForCausalLM.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="np")
>>> outputs = model(**inputs)

>>> # retrieve logts for next token
>>> next_token_logits = outputs.logits[:, -1]
```

### FlaxRobertaForMaskedLM

#### class transformers.FlaxRobertaForMaskedLM

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1056)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

RoBERTa Model with a `language modeling` head on top.

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxBaseModelOutputWithPooling](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **last\_hidden\_state** (`jnp.ndarray` of shape `(batch_size, sequence_length, hidden_size)`) — Sequence of hidden-states at the output of the last layer of the model.
* **pooler\_output** (`jnp.ndarray` of shape `(batch_size, hidden_size)`) — Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForMaskedLM

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForMaskedLM.from_pretrained("roberta-base")

>>> inputs = tokenizer("The capital of France is [MASK].", return_tensors="jax")

>>> outputs = model(**inputs)
>>> logits = outputs.logits
```

### FlaxRobertaForSequenceClassification

#### class transformers.FlaxRobertaForSequenceClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1128)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

Roberta 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 [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxSequenceClassifierOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxSequenceClassifierOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxSequenceClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxSequenceClassifierOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **logits** (`jnp.ndarray` of shape `(batch_size, config.num_labels)`) — Classification (or regression if config.num\_labels==1) scores (before SoftMax).
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForSequenceClassification

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForSequenceClassification.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="jax")

>>> outputs = model(**inputs)
>>> logits = outputs.logits
```

### FlaxRobertaForMultipleChoice

#### class transformers.FlaxRobertaForMultipleChoice

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1209)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

Roberta Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a softmax) e.g. for RocStories/SWAG tasks.

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxMultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, num_choices, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` of shape `(batch_size, num_choices, 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` of shape `(batch_size, num_choices, 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, num_choices, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxMultipleChoiceModelOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxMultipleChoiceModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxMultipleChoiceModelOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **logits** (`jnp.ndarray` of shape `(batch_size, num_choices)`) — *num\_choices* is the second dimension of the input tensors. (see *input\_ids* above).

  Classification scores (before SoftMax).
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForMultipleChoice

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForMultipleChoice.from_pretrained("roberta-base")

>>> prompt = "In Italy, pizza served in formal settings, such as at a restaurant, is presented unsliced."
>>> choice0 = "It is eaten with a fork and a knife."
>>> choice1 = "It is eaten while held in the hand."

>>> encoding = tokenizer([prompt, prompt], [choice0, choice1], return_tensors="jax", padding=True)
>>> outputs = model(**{k: v[None, :] for k, v in encoding.items()})

>>> logits = outputs.logits
```

### FlaxRobertaForTokenClassification

#### class transformers.FlaxRobertaForTokenClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1291)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

Roberta 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.

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxTokenClassifierOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxTokenClassifierOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxTokenClassifierOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxTokenClassifierOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **logits** (`jnp.ndarray` of shape `(batch_size, sequence_length, config.num_labels)`) — Classification scores (before SoftMax).
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForTokenClassification

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForTokenClassification.from_pretrained("roberta-base")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="jax")

>>> outputs = model(**inputs)
>>> logits = outputs.logits
```

### FlaxRobertaForQuestionAnswering

#### class transformers.FlaxRobertaForQuestionAnswering

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L1368)

( config: RobertaConfiginput\_shape: typing.Tuple = (1, 1)seed: int = 0dtype: dtype = \<class 'jax.numpy.float32'>\_do\_init: bool = Truegradient\_checkpointing: bool = False\*\*kwargs )

Parameters

* **config** ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method to load the model weights.

Roberta 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`).

This model inherits from [FlaxPreTrainedModel](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.FlaxPreTrainedModel). Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)

This model is also a Flax Linen [flax.linen.Module](https://flax.readthedocs.io/en/latest/flax.linen.html#module) subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.

Finally, this model supports inherent JAX features such as:

* [Just-In-Time (JIT) compilation](https://jax.readthedocs.io/en/latest/jax.html#just-in-time-compilation-jit)
* [Automatic Differentiation](https://jax.readthedocs.io/en/latest/jax.html#automatic-differentiation)
* [Vectorization](https://jax.readthedocs.io/en/latest/jax.html#vectorization-vmap)
* [Parallelization](https://jax.readthedocs.io/en/latest/jax.html#parallelization-pmap)

**\_\_call\_\_**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/roberta/modeling_flax_roberta.py#L816)

( input\_idsattention\_mask = Nonetoken\_type\_ids = Noneposition\_ids = Nonehead\_mask = Noneencoder\_hidden\_states = Noneencoder\_attention\_mask = Noneparams: dict = Nonedropout\_rng: PRNGKey = Nonetrain: bool = Falseoutput\_attentions: typing.Optional\[bool] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = Nonepast\_key\_values: dict = None ) → [transformers.modeling\_flax\_outputs.FlaxQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxQuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`

Parameters

* **input\_ids** (`numpy.ndarray` of shape `(batch_size, sequence_length)`) — Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using [AutoTokenizer](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/tokenizer#transformers.PreTrainedTokenizerFast.encode) and [PreTrainedTokenizer.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/vits#transformers.VitsTokenizer.__call__) for details.

  [What are input IDs?](https://huggingface.co/docs/transformers/glossary#input-ids)
* **attention\_mask** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#attention-mask)
* **token\_type\_ids** (`numpy.ndarray` 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?](https://huggingface.co/docs/transformers/glossary#token-type-ids)
* **position\_ids** (`numpy.ndarray` 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]`.
* **head\_mask** (`numpy.ndarray` of shape `(batch_size, sequence_length)`, `optional) -- Mask to nullify selected heads of the attention modules. Mask values selected in` \[0, 1]\`:
  * 1 indicates the head is **not masked**,
  * 0 indicates the head is **masked**.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

[transformers.modeling\_flax\_outputs.FlaxQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxQuestionAnsweringModelOutput) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_flax\_outputs.FlaxQuestionAnsweringModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_flax_outputs.FlaxQuestionAnsweringModelOutput) 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 ([RobertaConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/roberta#transformers.RobertaConfig)) and inputs.

* **start\_logits** (`jnp.ndarray` of shape `(batch_size, sequence_length)`) — Span-start scores (before SoftMax).
* **end\_logits** (`jnp.ndarray` of shape `(batch_size, sequence_length)`) — Span-end scores (before SoftMax).
* **hidden\_states** (`tuple(jnp.ndarray)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) — Tuple of `jnp.ndarray` (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(jnp.ndarray)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) — Tuple of `jnp.ndarray` (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 `FlaxRobertaPreTrainedModel` 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:

Copied

```
>>> from transformers import AutoTokenizer, FlaxRobertaForQuestionAnswering

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = FlaxRobertaForQuestionAnswering.from_pretrained("roberta-base")

>>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
>>> inputs = tokenizer(question, text, return_tensors="jax")

>>> outputs = model(**inputs)
>>> start_scores = outputs.start_logits
>>> end_scores = outputs.end_logits
```