RoBERTa

Overview

The RoBERTa model was proposed in RoBERTa: A Robustly Optimized BERT Pretraining Approach by Yinhan Liu, Myle Ott, 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 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 pre-training 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>)

• Camembert is a wrapper around RoBERTa. Refer to this page for usage examples.

The original code can be found here.

RobertaConfig

class transformers.RobertaConfig(pad_token_id=1, bos_token_id=0, eos_token_id=2, **kwargs)

This is the configuration class to store the configuration of a RobertaModel. It is used to instantiate an RoBERTa model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the BERT bert-base-uncased architecture.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

The RobertaConfig class directly inherits BertConfig. It reuses the same defaults. Please check the parent class for more information.

Example:

>>> from transformers import RobertaConfig, RobertaModel

>>> # Initializing a RoBERTa configuration
>>> configuration = RobertaConfig()

>>> # Initializing a model from the configuration
>>> model = RobertaModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

RobertaTokenizer

class transformers.RobertaTokenizer(vocab_file, merges_file, errors='replace', bos_token='<s>', eos_token='</s>', sep_token='</s>', cls_token='<s>', unk_token='<unk>', pad_token='<pad>', mask_token='<mask>', add_prefix_space=False, **kwargs)

Constructs a RoBERTa BPE tokenizer, derived from the GPT-2 tokenizer, using byte-level Byte-Pair-Encoding.

This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not:

>>> from transformers import RobertaTokenizer
>>> tokenizer = RobertaTokenizer.from_pretrained("roberta-base")
>>> tokenizer("Hello world")['input_ids']
[0, 31414, 232, 328, 2]
>>> tokenizer(" Hello world")['input_ids']
[0, 20920, 232, 2]

You can get around that behavior by passing add_prefix_space=True when instantiating this tokenizer or when you call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.

Note

When used with is_pretokenized=True, this tokenizer will add a space before each word (even the first one).

This tokenizer inherits from PreTrainedTokenizer which contains most of the methods. Users should refer to the superclass for more information regarding methods.

Parameters

• vocab_file (str) – Path to the vocabulary file.

• merges_file (str) – Path to the merges file.

• errors (str, optional, defaults to “replace”) – Paradigm to follow when decoding bytes to UTF-8. See bytes.decode for more information.

• bos_token (string, optional, defaults to “<s>”) –

  The beginning of sequence token that was used during pre-training. Can be used a sequence classifier token.

  Note

  When building a sequence using special tokens, this is not the token that is used for the beginning of sequence. The token used is the cls_token.

• eos_token (string, optional, defaults to “</s>”) –

  The end of sequence token.

  Note

  When building a sequence using special tokens, this is not the token that is used for the end of sequence. The token used is the sep_token.

• sep_token (string, optional, defaults to “</s>”) – The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens.

• cls_token (string, optional, defaults to “<s>”) – The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens.

• unk_token (string, optional, defaults to “<unk>”) – The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.

• pad_token (string, optional, defaults to “<pad>”) – The token used for padding, for example when batching sequences of different lengths.

• mask_token (string, optional, defaults to “<mask>”) – The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict.

build_inputs_with_special_tokens(token_ids_0: List[int], token_ids_1: Optional[List[int]] = None) → List[int]

Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. A RoBERTa sequence has the following format:

• single sequence: <s> X </s>

• pair of sequences: <s> A </s></s> B </s>

Parameters

• token_ids_0 (List[int]) – List of IDs to which the special tokens will be added

• token_ids_1 (List[int], optional, defaults to None) – Optional second list of IDs for sequence pairs.

Returns

list of input IDs with the appropriate special tokens.

Return type

List[int]

create_token_type_ids_from_sequences(token_ids_0: List[int], token_ids_1: Optional[List[int]] = None) → List[int]

Creates a mask from the two sequences passed to be used in a sequence-pair classification task. RoBERTa does not make use of token type ids, therefore a list of zeros is returned.

Parameters

• token_ids_0 (List[int]) – List of ids.

• token_ids_1 (List[int], optional, defaults to None) – Optional second list of IDs for sequence pairs.

Returns

List of zeros.

Return type

List[int]

get_special_tokens_mask(token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False) → List[int]

Retrieves sequence ids from a token list that has no special tokens added. This method is called when adding special tokens using the tokenizer prepare_for_model method.

Parameters

• token_ids_0 (List[int]) – List of ids.

• token_ids_1 (List[int], optional, defaults to None) – Optional second list of IDs for sequence pairs.

• already_has_special_tokens (bool, optional, defaults to False) – Set to True if the token list is already formatted with special tokens for the model

Returns

A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.

Return type

List[int]

save_vocabulary(save_directory)

Save the vocabulary and special tokens file to a directory.

Parameters

save_directory (str) – The directory in which to save the vocabulary.

Returns

Paths to the files saved.

Return type

Tuple(str)

RobertaTokenizerFast

class transformers.RobertaTokenizerFast(vocab_file, merges_file, errors='replace', bos_token='<s>', eos_token='</s>', sep_token='</s>', cls_token='<s>', unk_token='<unk>', pad_token='<pad>', mask_token='<mask>', add_prefix_space=False, trim_offsets=True, **kwargs)

Constructs a “Fast” RoBERTa BPE tokenizer (backed by HuggingFace’s tokenizers library), derived from the GPT-2 tokenizer, using byte-level Byte-Pair-Encoding.

This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not:

>>> from transformers import RobertaTokenizerFast
>>> tokenizer = RobertaTokenizerFast.from_pretrained("roberta-base")
>>> tokenizer("Hello world")['input_ids']
[0, 31414, 232, 328, 2]
>>> tokenizer(" Hello world")['input_ids']
[0, 20920, 232, 2]

You can get around that behavior by passing add_prefix_space=True when instantiating this tokenizer or when you call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.

Note

When used with is_pretokenized=True, this tokenizer needs to be instantiated with add_prefix_space=True.

This tokenizer inherits from PreTrainedTokenizerFast which contains most of the methods. Users should refer to the superclass for more information regarding methods.

Parameters

• vocab_file (str) – Path to the vocabulary file.

• merges_file (str) – Path to the merges file.

• errors (str, optional, defaults to “replace”) – Paradigm to follow when decoding bytes to UTF-8. See bytes.decode for more information.

• unk_token (string, optional, defaults to <|endoftext|>) – The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.

• bos_token (string, optional, defaults to <|endoftext|>) – The beginning of sequence token.

• eos_token (string, optional, defaults to <|endoftext|>) – The end of sequence token.

• add_prefix_space (bool, optional, defaults to False) – Whether to add a leading space to the first word. This allows to treat the leading word just as any other word. (GPT2 tokenizer detect beginning of words by the preceeding space)

• trim_offsets (bool, optional, defaults to True) – Whether the post processing step should trim offsets to avoid including whitespaces.

build_inputs_with_special_tokens(token_ids_0, token_ids_1=None)

Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. This implementation does not add special tokens.

RobertaModel

class transformers.RobertaModel(config)

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

This model is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

This class overrides BertModel. Please check the superclass for the appropriate documentation alongside usage examples.

config_class

alias of transformers.configuration_roberta.RobertaConfig

get_input_embeddings()

Returns the model’s input embeddings.

Returns

A torch module mapping vocabulary to hidden states.

Return type

nn.Module

set_input_embeddings(value)

Set model’s input embeddings

Parameters

value (nn.Module) – A module mapping vocabulary to hidden states.

RobertaForMaskedLM

class transformers.RobertaForMaskedLM(config)

RoBERTa Model with a language modeling head on top.

This model is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

config_class

alias of transformers.configuration_roberta.RobertaConfig

forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, output_attentions=None, output_hidden_states=None, **kwargs)

The RobertaForMaskedLM forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) –

  Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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, defaults to None) – 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, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) – 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

masked_lm_loss (optional, returned when labels is provided) torch.FloatTensor of shape (1,):

Masked language modeling loss.

prediction_scores (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 + 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(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.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, RobertaForMaskedLM
>>> import torch

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = RobertaForMaskedLM.from_pretrained('roberta-base')

>>> input_ids = tokenizer("Hello, my dog is cute", return_tensors="pt")["input_ids"]

>>> outputs = model(input_ids, labels=input_ids)
>>> loss, prediction_scores = outputs[:2]

get_output_embeddings()

Returns the model’s output embeddings.

Returns

A torch module mapping hidden states to vocabulary.

Return type

nn.Module

RobertaForSequenceClassification

class transformers.RobertaForSequenceClassification(config)

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 is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

config_class

alias of transformers.configuration_roberta.RobertaConfig

forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, output_attentions=None, output_hidden_states=None)

The RobertaForSequenceClassification forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) –

  Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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, defaults to None) – 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, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (torch.LongTensor of shape (batch_size,), optional, defaults to None) – 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

loss (torch.FloatTensor of shape (1,), optional, returned when label 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 + 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(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.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, RobertaForSequenceClassification
>>> import torch

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = RobertaForSequenceClassification.from_pretrained('roberta-base')

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> labels = torch.tensor([1]).unsqueeze(0)  # Batch size 1
>>> outputs = model(**inputs, labels=labels)
>>> loss, logits = outputs[:2]

RobertaForMultipleChoice

class transformers.RobertaForMultipleChoice(config)

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 is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

config_class

alias of transformers.configuration_roberta.RobertaConfig

forward(input_ids=None, token_type_ids=None, attention_mask=None, labels=None, position_ids=None, head_mask=None, inputs_embeds=None, output_attentions=None, output_hidden_states=None)

The RobertaForMultipleChoice forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, num_choices, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (torch.LongTensor of shape (batch_size, num_choices, sequence_length), optional, defaults to None) –

  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?

• position_ids (torch.LongTensor of shape (batch_size, num_choices, sequence_length), optional, defaults to None) –

  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?

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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, defaults to None) – 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, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (torch.LongTensor of shape (batch_size,), optional, defaults to None) – 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

loss (torch.FloatTensor` of shape (1,), optional, returned when labels is provided):

Classification loss.

classification_scores (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 + 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(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.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, RobertaForMultipleChoice
>>> import torch

>>> tokenizer = RobertaTokenizer.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, logits = outputs[:2]

RobertaForTokenClassification

class transformers.RobertaForTokenClassification(config)

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 is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

config_class

alias of transformers.configuration_roberta.RobertaConfig

forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, labels=None, output_attentions=None, output_hidden_states=None)

The RobertaForTokenClassification forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) –

  Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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, defaults to None) – 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, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) – Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

Returns

loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) :

Classification loss.

scores (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 + 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(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.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, RobertaForTokenClassification
>>> import torch

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = RobertaForTokenClassification.from_pretrained('roberta-base')

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> labels = torch.tensor([1] * inputs["input_ids"].size(1)).unsqueeze(0)  # Batch size 1

>>> outputs = model(**inputs, labels=labels)
>>> loss, scores = outputs[:2]

RobertaForQuestionAnswering

class transformers.RobertaForQuestionAnswering(config)

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 is a PyTorch torch.nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

Parameters

config (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() method to load the model weights.

config_class

alias of transformers.configuration_roberta.RobertaConfig

forward(input_ids=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, start_positions=None, end_positions=None, output_attentions=None, output_hidden_states=None)

The RobertaForQuestionAnswering forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) –

  Indices of input sequence tokens in the vocabulary.

  Indices can be obtained using transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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, defaults to None) – 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, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• start_positions (torch.LongTensor of shape (batch_size,), optional, defaults to None) – 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, defaults to None) – 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

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_scores (torch.FloatTensor of shape (batch_size, sequence_length,)):

Span-start scores (before SoftMax).

end_scores (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 + 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(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.

Return type

tuple(torch.FloatTensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, RobertaForQuestionAnswering
>>> import torch

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = RobertaForQuestionAnswering.from_pretrained('roberta-base')

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> start_positions = torch.tensor([1])
>>> end_positions = torch.tensor([3])

>>> outputs = model(**inputs, start_positions=start_positions, end_positions=end_positions)
>>> loss, start_scores, end_scores = outputs[:3]

TFRobertaModel

class transformers.TFRobertaModel(*args, **kwargs)

The bare RoBERTa Model transformer outputing raw hidden-states without any specific head on top. This model is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs, **kwargs)

The TFRobertaModel forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

Returns

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 Bert 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.

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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, TFRobertaModel
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.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[0]  # The last hidden-state is the first element of the output tuple

TFRobertaForMaskedLM

class transformers.TFRobertaForMaskedLM(*args, **kwargs)

RoBERTa Model with a language modeling head on top. This model is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs, **kwargs)

The TFRobertaForMaskedLM forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

Returns

prediction_scores (Numpy array or 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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example::

>>> from transformers import RobertaTokenizer, TFRobertaForMaskedLM
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = TFRobertaForMaskedLM.from_pretrained('roberta-base')

>>> input_ids = tf.constant(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True))[None, :]  # Batch size 1

>>> outputs = model(input_ids)
>>> prediction_scores = outputs[0]

get_output_embeddings()

Returns the model’s output embeddings.

Returns

A torch module mapping hidden states to vocabulary.

Return type

tf.keras.layers.Layer

TFRobertaForSequenceClassification

class transformers.TFRobertaForSequenceClassification(*args, **kwargs)

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 is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, output_attentions=None, output_hidden_states=None, labels=None, training=False)

The TFRobertaForSequenceClassification forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

Returns

logits (Numpy array or 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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, TFRobertaForSequenceClassification
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = TFRobertaForSequenceClassification.from_pretrained('roberta-base')

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf")
>>> inputs["labels"] = tf.reshape(tf.constant(1), (-1, 1)) # Batch size 1

>>> outputs = model(inputs)
>>> loss, logits = outputs[:2]

TFRobertaForMultipleChoice

class transformers.TFRobertaForMultipleChoice(*args, **kwargs)

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 is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, output_attentions=None, output_hidden_states=None, labels=None, training=False)

The TFRobertaForMultipleChoice forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (tf.Tensor of shape (batch_size,), optional, defaults to None) – 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

classification_scores (Numpy array or tf.Tensor of shape (batch_size, num_choices):

num_choices is the size of 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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (BertConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, TFRobertaForMultipleChoice
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.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[0]

property dummy_inputs

Dummy inputs to build the network.

Returns

tf.Tensor with dummy inputs

TFRobertaForTokenClassification

class transformers.TFRobertaForTokenClassification(*args, **kwargs)

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 is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, output_attentions=None, output_hidden_states=None, labels=None, training=False)

The TFRobertaForTokenClassification forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• labels (tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) – Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

Returns

scores (Numpy array or 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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, TFRobertaForTokenClassification
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = TFRobertaForTokenClassification.from_pretrained('roberta-base')

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="tf")
>>> input_ids = inputs["input_ids"]
>>> inputs["labels"] = tf.reshape(tf.constant([1] * tf.size(input_ids).numpy()), (-1, tf.size(input_ids))) # Batch size 1

>>> outputs = model(inputs)
>>> loss, scores = outputs[:2]

TFRobertaForQuestionAnswering

class transformers.TFRobertaForQuestionAnswering(*args, **kwargs)

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 is a tf.keras.Model sub-class. 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.

Note

TF 2.0 models accepts two formats as inputs:

• having all inputs as keyword arguments (like PyTorch models), or

• having all inputs as a list, tuple or dict in the first positional arguments.

This second option is useful when using tf.keras.Model.fit() method which currently requires having all the tensors in the first argument of the model call function: model(inputs).

If you choose this second option, 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(inputs_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})

Parameters

config (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() method to load the model weights.

call(inputs=None, attention_mask=None, token_type_ids=None, position_ids=None, head_mask=None, inputs_embeds=None, output_attentions=None, output_hidden_states=None, start_positions=None, end_positions=None, training=False)

The TFRobertaForQuestionAnswering forward method, overrides the __call__() special method.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Parameters

• 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 transformers.RobertaTokenizer. See transformers.PreTrainedTokenizer.encode() and transformers.PreTrainedTokenizer.__call__() for details.

  What are input IDs?

• attention_mask (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]: 1 for tokens that are NOT MASKED, 0 for MASKED tokens.

  What are attention masks?

• token_type_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• position_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length), optional, defaults to None) –

  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?

• head_mask (Numpy array or tf.Tensor of shape (num_heads,) or (num_layers, num_heads), optional, defaults to None) – 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 (Numpy array or tf.Tensor of shape (batch_size, sequence_length, embedding_dim), optional, defaults to None) – 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.

• training (boolean, optional, defaults to False) – Whether to activate dropout modules (if set to True) during training or to de-activate them (if set to False) for evaluation.

• output_attentions (bool, optional, defaults to None) – If set to True, the attentions tensors of all attention layers are returned. See attentions under returned tensors for more detail.

• start_positions (tf.Tensor of shape (batch_size,), optional, defaults to None) – 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, defaults to None) – 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

start_scores (Numpy array or tf.Tensor of shape (batch_size, sequence_length,)):

Span-start scores (before SoftMax).

end_scores (Numpy array or 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.

Return type

tuple(tf.Tensor) comprising various elements depending on the configuration (RobertaConfig) and inputs

Example:

>>> from transformers import RobertaTokenizer, TFRobertaForQuestionAnswering
>>> import tensorflow as tf

>>> tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
>>> model = TFRobertaForQuestionAnswering.from_pretrained('roberta-base')

>>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
>>> input_dict = tokenizer(question, text, return_tensors='tf')
>>> start_scores, end_scores = model(input_dict)

>>> all_tokens = tokenizer.convert_ids_to_tokens(input_dict["input_ids"].numpy()[0])
>>> answer = ' '.join(all_tokens[tf.math.argmax(start_scores, 1)[0] : tf.math.argmax(end_scores, 1)[0]+1])