This model was released on 2019-07-26 and added to Hugging Face Transformers on 2020-11-16.
RoBERTa
RoBERTa improves BERT with new pretraining objectives, demonstratingBERT was undertrained and training design is important. The pretraining objectives include dynamic masking, sentence packing, larger batches and a byte-level BPE tokenizer.
You can find all the original RoBERTa checkpoints under theFacebook AI organization.
Click on the RoBERTa models in the right sidebar for more examples of how to apply RoBERTa to different language tasks.
The example below demonstrates how to predict the<mask> token withPipeline,AutoModel, and from the command line.
import torchfrom transformersimport pipelinepipeline = pipeline( task="fill-mask", model="FacebookAI/roberta-base", dtype=torch.float16, device=0)pipeline("Plants create <mask> through a process known as photosynthesis.")
Notes
- RoBERTa doesn’t have
token_type_idsso you don’t need to indicate which token belongs to which segment. Separate your segments with the separation tokentokenizer.sep_tokenor</s>.
RobertaConfig
classtransformers.RobertaConfig
<source>(vocab_size = 50265hidden_size = 768num_hidden_layers = 12num_attention_heads = 12intermediate_size = 3072hidden_act = 'gelu'hidden_dropout_prob = 0.1attention_probs_dropout_prob = 0.1max_position_embeddings = 512type_vocab_size = 2initializer_range = 0.02layer_norm_eps = 1e-12pad_token_id = 1bos_token_id = 0eos_token_id = 2use_cache = Trueclassifier_dropout = None**kwargs)
Parameters
- vocab_size (
int,optional, defaults to 50265) —Vocabulary size of the RoBERTa model. Defines the number of different tokens that can be represented by theinputs_idspassed when callingRobertaModel orTFRobertaModel. - hidden_size (
int,optional, defaults to 768) —Dimensionality of the encoder layers and the pooler layer. - num_hidden_layers (
int,optional, defaults to 12) —Number of hidden layers in the Transformer encoder. - num_attention_heads (
int,optional, defaults to 12) —Number of attention heads for each attention layer in the Transformer encoder. - intermediate_size (
int,optional, defaults to 3072) —Dimensionality of the “intermediate” (often named feed-forward) layer in the Transformer encoder. - hidden_act (
strorCallable,optional, defaults to"gelu") —The non-linear activation function (function or string) in the encoder and pooler. If string,"gelu","relu","silu"and"gelu_new"are supported. - hidden_dropout_prob (
float,optional, defaults to 0.1) —The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. - attention_probs_dropout_prob (
float,optional, defaults to 0.1) —The dropout ratio for the attention probabilities. - max_position_embeddings (
int,optional, defaults to 512) —The maximum sequence length that this model might ever be used with. Typically set this to something largejust in case (e.g., 512 or 1024 or 2048). - type_vocab_size (
int,optional, defaults to 2) —The vocabulary size of thetoken_type_idspassed when callingRobertaModel orTFRobertaModel. - initializer_range (
float,optional, defaults to 0.02) —The standard deviation of the truncated_normal_initializer for initializing all weight matrices. - layer_norm_eps (
float,optional, defaults to 1e-12) —The epsilon used by the layer normalization layers. - is_decoder (
bool,optional, defaults toFalse) —Whether the model is used as a decoder or not. IfFalse, the model is used as an encoder. - use_cache (
bool,optional, defaults toTrue) —Whether or not the model should return the last key/values attentions (not used by all models). Onlyrelevant ifconfig.is_decoder=True. - classifier_dropout (
float,optional) —The dropout ratio for the classification head.
This is the configuration class to store the configuration of aRobertaModel or aTFRobertaModel. It isused to instantiate a 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 RoBERTaFacebookAI/roberta-base architecture.
Configuration objects inherit fromPreTrainedConfig and can be used to control the model outputs. Read thedocumentation fromPreTrainedConfig for more information.
Examples:
>>>from transformersimport RobertaConfig, RobertaModel>>># Initializing a RoBERTa configuration>>>configuration = RobertaConfig()>>># Initializing a model (with random weights) from the configuration>>>model = RobertaModel(configuration)>>># Accessing the model configuration>>>configuration = model.config
RobertaTokenizer
classtransformers.RobertaTokenizer
<source>(vocab_filemerges_fileerrors = '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)
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. Seebytes.decode for more information. - bos_token (
str,optional, defaults to"<s>") —The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.When building a sequence using special tokens, this is not the token that is used for the beginning ofsequence. The token used is the
cls_token. - eos_token (
str,optional, defaults to"</s>") —The end of sequence token.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 (
str,optional, defaults to"</s>") —The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences forsequence classification or for a text and a question for question answering. It is also used as the lasttoken of a sequence built with special tokens. - cls_token (
str,optional, defaults to"<s>") —The classifier token which is used when doing sequence classification (classification of the whole sequenceinstead of per-token classification). It is the first token of the sequence when built with special tokens. - unk_token (
str,optional, defaults to"<unk>") —The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be thistoken instead. - pad_token (
str,optional, defaults to"<pad>") —The token used for padding, for example when batching sequences of different lengths. - mask_token (
str,optional, defaults to"<mask>") —The token used for masking values. This is the token used when training this model with masked languagemodeling. This is the token which the model will try to predict. - add_prefix_space (
bool,optional, defaults toFalse) —Whether or not to add an initial space to the input. This allows to treat the leading word just as anyother word. (RoBERTa tokenizer detect beginning of words by the preceding space).
Constructs a RoBERTa 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 transformersimport RobertaTokenizer>>>tokenizer = RobertaTokenizer.from_pretrained("FacebookAI/roberta-base")>>>tokenizer("Hello world")["input_ids"][0,31414,232,2]>>>tokenizer(" Hello world")["input_ids"][0,20920,232,2]
You can get around that behavior by passingadd_prefix_space=True when instantiating this tokenizer or when youcall it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.
When used with
is_split_into_words=True, this tokenizer will add a space before each word (even the first one).
This tokenizer inherits fromPreTrainedTokenizer which contains most of the main methods. Users should refer tothis superclass for more information regarding those methods.
build_inputs_with_special_tokens
<source>(token_ids_0: listtoken_ids_1: typing.Optional[list[int]] = None)→list[int]
Parameters
- token_ids_0 (
list[int]) —List of IDs to which the special tokens will be added. - token_ids_1 (
list[int],optional) —Optional second list of IDs for sequence pairs.
Returns
list[int]
List ofinput IDs with the appropriate special tokens.
Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating andadding special tokens. A RoBERTa sequence has the following format:
- single sequence:
<s> X </s> - pair of sequences:
<s> A </s></s> B </s>
get_special_tokens_mask
<source>(token_ids_0: listtoken_ids_1: typing.Optional[list[int]] = Nonealready_has_special_tokens: bool = False)→list[int]
Parameters
- token_ids_0 (
list[int]) —List of IDs. - token_ids_1 (
list[int],optional) —Optional second list of IDs for sequence pairs. - already_has_special_tokens (
bool,optional, defaults toFalse) —Whether or not the token list is already formatted with special tokens for the model.
Returns
list[int]
A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
Retrieve sequence ids from a token list that has no special tokens added. This method is called when addingspecial tokens using the tokenizerprepare_for_model method.
create_token_type_ids_from_sequences
<source>(token_ids_0: listtoken_ids_1: typing.Optional[list[int]] = None)→list[int]
Create a mask from the two sequences passed to be used in a sequence-pair classification task. RoBERTa does notmake use of token type ids, therefore a list of zeros is returned.
save_vocabulary
<source>(save_directory: strfilename_prefix: typing.Optional[str] = None)
RobertaTokenizerFast
classtransformers.RobertaTokenizerFast
<source>(vocab_file = Nonemerges_file = Nonetokenizer_file = Noneerrors = '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 = Falsetrim_offsets = True**kwargs)
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. Seebytes.decode for more information. - bos_token (
str,optional, defaults to"<s>") —The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.When building a sequence using special tokens, this is not the token that is used for the beginning ofsequence. The token used is the
cls_token. - eos_token (
str,optional, defaults to"</s>") —The end of sequence token.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 (
str,optional, defaults to"</s>") —The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences forsequence classification or for a text and a question for question answering. It is also used as the lasttoken of a sequence built with special tokens. - cls_token (
str,optional, defaults to"<s>") —The classifier token which is used when doing sequence classification (classification of the whole sequenceinstead of per-token classification). It is the first token of the sequence when built with special tokens. - unk_token (
str,optional, defaults to"<unk>") —The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be thistoken instead. - pad_token (
str,optional, defaults to"<pad>") —The token used for padding, for example when batching sequences of different lengths. - mask_token (
str,optional, defaults to"<mask>") —The token used for masking values. This is the token used when training this model with masked languagemodeling. This is the token which the model will try to predict. - add_prefix_space (
bool,optional, defaults toFalse) —Whether or not to add an initial space to the input. This allows to treat the leading word just as anyother word. (RoBERTa tokenizer detect beginning of words by the preceding space). - trim_offsets (
bool,optional, defaults toTrue) —Whether the post processing step should trim offsets to avoid including whitespaces.
Construct a “fast” RoBERTa tokenizer (backed by HuggingFace’stokenizers library), derived from the GPT-2tokenizer, 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 transformersimport RobertaTokenizerFast>>>tokenizer = RobertaTokenizerFast.from_pretrained("FacebookAI/roberta-base")>>>tokenizer("Hello world")["input_ids"][0,31414,232,2]>>>tokenizer(" Hello world")["input_ids"][0,20920,232,2]
You can get around that behavior by passingadd_prefix_space=True when instantiating this tokenizer or when youcall it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.
When used with
is_split_into_words=True, this tokenizer needs to be instantiated withadd_prefix_space=True.
This tokenizer inherits fromPreTrainedTokenizerFast which contains most of the main methods. Users shouldrefer to this superclass for more information regarding those methods.
build_inputs_with_special_tokens
<source>(token_ids_0token_ids_1 = None)
RobertaModel
classtransformers.RobertaModel
<source>(configadd_pooling_layer = True)
Parameters
- config (RobertaModel) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
- add_pooling_layer (
bool,optional, defaults toTrue) —Whether to add a pooling layer
The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer ofcross-attention is added between the self-attention layers, following the architecture described inAttention isall you need by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
To behave as an decoder the model needs to be initialized with theis_decoder argument of the configuration settoTrue. To be used in a Seq2Seq model, the model needs to initialized with bothis_decoder argument andadd_cross_attention set toTrue; anencoder_hidden_states is then expected as an input to the forward pass.
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(input_ids: typing.Optional[torch.Tensor] = Noneattention_mask: typing.Optional[torch.Tensor] = Nonetoken_type_ids: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.Tensor] = Noneinputs_embeds: typing.Optional[torch.Tensor] = Noneencoder_hidden_states: typing.Optional[torch.Tensor] = Noneencoder_attention_mask: typing.Optional[torch.Tensor] = Nonepast_key_values: typing.Optional[transformers.cache_utils.Cache] = Noneuse_cache: typing.Optional[bool] = Nonecache_position: typing.Optional[torch.Tensor] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.BaseModelOutputWithPoolingAndCrossAttentions ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.Tensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.Tensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.Tensorof 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 asentence A token,
- 1 corresponds to asentence B token.
- position_ids (
torch.Tensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.Tensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - encoder_hidden_states (
torch.Tensorof 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-attentionif the model is configured as a decoder. - encoder_attention_mask (
torch.Tensorof shape(batch_size, sequence_length),optional) —Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used inthe cross-attention if the model is configured as a decoder. Mask values selected in[0, 1]:- 1 for tokens that arenot masked,
- 0 for tokens that aremasked.
- past_key_values (
~cache_utils.Cache,optional) —Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attentionblocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_valuesreturned by the model at a previous stage of decoding, whenuse_cache=Trueorconfig.use_cache=True.OnlyCache instance is allowed as input, see ourkv cache guide.If no
past_key_valuesare passed,DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_valuesare used, the user is expected to input only unprocessedinput_ids(those that don’thave their past key value states given to this model) of shape(batch_size, unprocessed_length)instead of allinput_idsof shape(batch_size, sequence_length). - use_cache (
bool,optional) —If set toTrue,past_key_valueskey value states are returned and can be used to speed up decoding (seepast_key_values). - cache_position (
torch.Tensorof shape(sequence_length),optional) —Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids,this tensor is not affected by padding. It is used to update the cache in the correct position and to inferthe complete sequence length.
Returns
transformers.modeling_outputs.BaseModelOutputWithPoolingAndCrossAttentions ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.BaseModelOutputWithPoolingAndCrossAttentions or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
last_hidden_state (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.pooler_output (
torch.FloatTensorof shape(batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processingthrough the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returnsthe classification token after processing through a linear layer and a tanh activation function. The linearlayer weights are trained from the next sentence prediction (classification) objective during pretraining.hidden_states (
tuple(torch.FloatTensor),optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
cross_attentions (
tuple(torch.FloatTensor),optional, returned whenoutput_attentions=Trueandconfig.add_cross_attention=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.FloatTensor(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 theweighted average in the cross-attention heads.
past_key_values (
Cache,optional, returned whenuse_cache=Trueis passed or whenconfig.use_cache=True) — It is aCache instance. For more details, see ourkv cache guide.Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if
config.is_encoder_decoder=Truein the cross-attention blocks) that can be used (seepast_key_valuesinput) to speed up sequential decoding.
TheRobertaModel forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
RobertaForCausalLM
classtransformers.RobertaForCausalLM
<source>(config)
Parameters
- config (RobertaForCausalLM) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
RoBERTa Model with alanguage modeling head on top for CLM fine-tuning.
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = 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.Optional[tuple[tuple[torch.FloatTensor]]] = Noneuse_cache: typing.Optional[bool] = Nonecache_position: typing.Optional[torch.Tensor] = Nonelogits_to_keep: typing.Union[int, torch.Tensor] = 0**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.CausalLMOutputWithCrossAttentions ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- position_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - encoder_hidden_states (
torch.FloatTensorof 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-attentionif the model is configured as a decoder. - encoder_attention_mask (
torch.FloatTensorof shape(batch_size, sequence_length),optional) —Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used inthe cross-attention if the model is configured as a decoder. Mask values selected in[0, 1]:- 1 for tokens that arenot masked,
- 0 for tokens that aremasked.
- labels (
torch.LongTensorof 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](seeinput_idsdocstring) Tokens with indices set to-100areignored (masked), the loss is only computed for the tokens with labels in[0, ..., config.vocab_size] - past_key_values (
tuple[tuple[torch.FloatTensor]],optional) —Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attentionblocks) that can be used to speed up sequential decoding. This typically consists in thepast_key_valuesreturned by the model at a previous stage of decoding, whenuse_cache=Trueorconfig.use_cache=True.OnlyCache instance is allowed as input, see ourkv cache guide.If no
past_key_valuesare passed,DynamicCache will be initialized by default.The model will output the same cache format that is fed as input.
If
past_key_valuesare used, the user is expected to input only unprocessedinput_ids(those that don’thave their past key value states given to this model) of shape(batch_size, unprocessed_length)instead of allinput_idsof shape(batch_size, sequence_length). - use_cache (
bool,optional) —If set toTrue,past_key_valueskey value states are returned and can be used to speed up decoding (seepast_key_values). - cache_position (
torch.Tensorof shape(sequence_length),optional) —Indices depicting the position of the input sequence tokens in the sequence. Contrarily toposition_ids,this tensor is not affected by padding. It is used to update the cache in the correct position and to inferthe complete sequence length. - logits_to_keep (
Union[int, torch.Tensor], defaults to0) —If anint, compute logits for the lastlogits_to_keeptokens. If0, calculate logits for allinput_ids(special case). Only last token logits are needed for generation, and calculating them only for thattoken can save memory, which becomes pretty significant for long sequences or large vocabulary size.If atorch.Tensor, must be 1D corresponding to the indices to keep in the sequence length dimension.This is useful when using packed tensor format (single dimension for batch and sequence length).
Returns
transformers.modeling_outputs.CausalLMOutputWithCrossAttentions ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.CausalLMOutputWithCrossAttentions or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Language modeling loss (for next-token prediction).logits (
torch.FloatTensorof 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 whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
cross_attentions (
tuple(torch.FloatTensor),optional, returned whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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 thecross-attention heads.
past_key_values (
Cache,optional, returned whenuse_cache=Trueis passed or whenconfig.use_cache=True) — It is aCache instance. For more details, see ourkv cache guide.Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_valuesinput) to speed up sequential decoding.
TheRobertaForCausalLM forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example:
>>>from transformersimport AutoTokenizer, RobertaForCausalLM, AutoConfig>>>import torch>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>config = AutoConfig.from_pretrained("FacebookAI/roberta-base")>>>config.is_decoder =True>>>model = RobertaForCausalLM.from_pretrained("FacebookAI/roberta-base", config=config)>>>inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")>>>outputs = model(**inputs)>>>prediction_logits = outputs.logits
RobertaForMaskedLM
classtransformers.RobertaForMaskedLM
<source>(config)
Parameters
- config (RobertaForMaskedLM) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
The Roberta Model with alanguage modeling head on top.”
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = 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] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.MaskedLMOutput ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- position_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - encoder_hidden_states (
torch.FloatTensorof 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-attentionif the model is configured as a decoder. - encoder_attention_mask (
torch.FloatTensorof shape(batch_size, sequence_length),optional) —Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used inthe cross-attention if the model is configured as a decoder. Mask values selected in[0, 1]:- 1 for tokens that arenot masked,
- 0 for tokens that aremasked.
- labels (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Labels for computing the masked language modeling loss. Indices should be in[-100, 0, ..., config.vocab_size](seeinput_idsdocstring) Tokens with indices set to-100are ignored (masked), theloss is only computed for the tokens with labels in[0, ..., config.vocab_size]
Returns
transformers.modeling_outputs.MaskedLMOutput ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.MaskedLMOutput or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Masked language modeling (MLM) loss.logits (
torch.FloatTensorof 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 whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
TheRobertaForMaskedLM forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example:
>>>from transformersimport AutoTokenizer, RobertaForMaskedLM>>>import torch>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForMaskedLM.from_pretrained("FacebookAI/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)...>>>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)...
RobertaForSequenceClassification
classtransformers.RobertaForSequenceClassification
<source>(config)
Parameters
- config (RobertaForSequenceClassification) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
RoBERTa Model transformer with a sequence classification/regression head on top (a linear layer on top of thepooled output) e.g. for GLUE tasks.
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Nonelabels: typing.Optional[torch.LongTensor] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.SequenceClassifierOutput ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- position_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - labels (
torch.LongTensorof shape(batch_size,),optional) —Labels for computing the sequence classification/regression loss. Indices should be in[0, ..., config.num_labels - 1]. Ifconfig.num_labels == 1a regression loss is computed (Mean-Square loss), Ifconfig.num_labels > 1a classification loss is computed (Cross-Entropy).
Returns
transformers.modeling_outputs.SequenceClassifierOutput ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.SequenceClassifierOutput or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Classification (or regression if config.num_labels==1) loss.logits (
torch.FloatTensorof shape(batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax).hidden_states (
tuple(torch.FloatTensor),optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
TheRobertaForSequenceClassification forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example of single-label classification:
>>>import torch>>>from transformersimport AutoTokenizer, RobertaForSequenceClassification>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForSequenceClassification.from_pretrained("FacebookAI/roberta-base")>>>inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")>>>with torch.no_grad():... logits = model(**inputs).logits>>>predicted_class_id = logits.argmax().item()>>>model.config.id2label[predicted_class_id]...>>># To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`>>>num_labels =len(model.config.id2label)>>>model = RobertaForSequenceClassification.from_pretrained("FacebookAI/roberta-base", num_labels=num_labels)>>>labels = torch.tensor([1])>>>loss = model(**inputs, labels=labels).loss>>>round(loss.item(),2)...
Example of multi-label classification:
>>>import torch>>>from transformersimport AutoTokenizer, RobertaForSequenceClassification>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForSequenceClassification.from_pretrained("FacebookAI/roberta-base", problem_type="multi_label_classification")>>>inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")>>>with torch.no_grad():... logits = model(**inputs).logits>>>predicted_class_ids = torch.arange(0, logits.shape[-1])[torch.sigmoid(logits).squeeze(dim=0) >0.5]>>># To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`>>>num_labels =len(model.config.id2label)>>>model = RobertaForSequenceClassification.from_pretrained(..."FacebookAI/roberta-base", num_labels=num_labels, problem_type="multi_label_classification"...)>>>labels = torch.sum(... torch.nn.functional.one_hot(predicted_class_ids[None, :].clone(), num_classes=num_labels), dim=1...).to(torch.float)>>>loss = model(**inputs, labels=labels).loss
RobertaForMultipleChoice
classtransformers.RobertaForMultipleChoice
<source>(config)
Parameters
- config (RobertaForMultipleChoice) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
The Roberta Model with a multiple choice classification head on top (a linear layer on top of the pooled output and asoftmax) e.g. for RocStories/SWAG tasks.
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.MultipleChoiceModelOutput ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, num_choices, sequence_length)) —Indices of input sequence tokens in the vocabulary.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- labels (
torch.LongTensorof shape(batch_size,),optional) —Labels for computing the multiple choice classification loss. Indices should be in[0, ..., num_choices-1]wherenum_choicesis the size of the second dimension of the input tensors. (Seeinput_idsabove) - position_ids (
torch.LongTensorof 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]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, num_choices, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix.
Returns
transformers.modeling_outputs.MultipleChoiceModelOutput ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.MultipleChoiceModelOutput or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Classification loss.logits (
torch.FloatTensorof shape(batch_size, num_choices)) —num_choices is the second dimension of the input tensors. (seeinput_ids above).Classification scores (before SoftMax).
hidden_states (
tuple(torch.FloatTensor),optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
TheRobertaForMultipleChoice forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example:
>>>from transformersimport AutoTokenizer, RobertaForMultipleChoice>>>import torch>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForMultipleChoice.from_pretrained("FacebookAI/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, vin encoding.items()}, labels=labels)# batch size is 1>>># the linear classifier still needs to be trained>>>loss = outputs.loss>>>logits = outputs.logits
RobertaForTokenClassification
classtransformers.RobertaForTokenClassification
<source>(config)
Parameters
- config (RobertaForTokenClassification) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
The Roberta transformer with a token classification head on top (a linear layer on top of the hidden-statesoutput) e.g. for Named-Entity-Recognition (NER) tasks.
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Nonelabels: typing.Optional[torch.LongTensor] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.TokenClassifierOutput ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- position_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - labels (
torch.LongTensorof 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 ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.TokenClassifierOutput or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Classification loss.logits (
torch.FloatTensorof shape(batch_size, sequence_length, config.num_labels)) — Classification scores (before SoftMax).hidden_states (
tuple(torch.FloatTensor),optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
TheRobertaForTokenClassification forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example:
>>>from transformersimport AutoTokenizer, RobertaForTokenClassification>>>import torch>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForTokenClassification.from_pretrained("FacebookAI/roberta-base")>>>inputs = tokenizer(..."HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="pt"...)>>>with torch.no_grad():... logits = model(**inputs).logits>>>predicted_token_class_ids = logits.argmax(-1)>>># Note that tokens are classified rather then input words which means that>>># there might be more predicted token classes than words.>>># Multiple token classes might account for the same word>>>predicted_tokens_classes = [model.config.id2label[t.item()]for tin predicted_token_class_ids[0]]>>>predicted_tokens_classes...>>>labels = predicted_token_class_ids>>>loss = model(**inputs, labels=labels).loss>>>round(loss.item(),2)...
RobertaForQuestionAnswering
classtransformers.RobertaForQuestionAnswering
<source>(config)
Parameters
- config (RobertaForQuestionAnswering) —Model configuration class with all the parameters of the model. Initializing with a config file does notload the weights associated with the model, only the configuration. Check out thefrom_pretrained() method to load the model weights.
The Roberta transformer with a span classification head on top for extractive question-answering tasks likeSQuAD (a linear layer on top of the hidden-states output to computespan start logits andspan end logits).
This model inherits fromPreTrainedModel. Check the superclass documentation for the generic methods thelibrary implements for all its model (such as downloading or saving, resizing the input embeddings, pruning headsetc.)
This model is also a PyTorchtorch.nn.Module subclass.Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usageand behavior.
forward
<source>(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] = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Nonestart_positions: typing.Optional[torch.LongTensor] = Noneend_positions: typing.Optional[torch.LongTensor] = None**kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs])→transformers.modeling_outputs.QuestionAnsweringModelOutput ortuple(torch.FloatTensor)
Parameters
- input_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.Indices can be obtained usingAutoTokenizer. SeePreTrainedTokenizer.encode() andPreTrainedTokenizer.call() for details.
- attention_mask (
torch.FloatTensorof 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 arenot masked,
- 0 for tokens that aremasked.
- token_type_ids (
torch.LongTensorof 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 asentence A token,
- 1 corresponds to asentence B token.This parameter can only be used when the model is initialized with
type_vocab_sizeparameter with value= 2. All the value in this tensor should be always < type_vocab_size.
- position_ids (
torch.LongTensorof shape(batch_size, sequence_length),optional) —Indices of positions of each input sequence tokens in the position embeddings. Selected in the range[0, config.n_positions - 1]. - inputs_embeds (
torch.FloatTensorof shape(batch_size, sequence_length, hidden_size),optional) —Optionally, instead of passinginput_idsyou can choose to directly pass an embedded representation. Thisis useful if you want more control over how to convertinput_idsindices into associated vectors than themodel’s internal embedding lookup matrix. - start_positions (
torch.LongTensorof 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 sequenceare not taken into account for computing the loss. - end_positions (
torch.LongTensorof 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 sequenceare not taken into account for computing the loss.
Returns
transformers.modeling_outputs.QuestionAnsweringModelOutput ortuple(torch.FloatTensor)
Atransformers.modeling_outputs.QuestionAnsweringModelOutput or a tuple oftorch.FloatTensor (ifreturn_dict=False is passed or whenconfig.return_dict=False) comprising variouselements depending on the configuration (RobertaConfig) and inputs.
loss (
torch.FloatTensorof shape(1,),optional, returned whenlabelsis provided) — Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.start_logits (
torch.FloatTensorof shape(batch_size, sequence_length)) — Span-start scores (before SoftMax).end_logits (
torch.FloatTensorof shape(batch_size, sequence_length)) — Span-end scores (before SoftMax).hidden_states (
tuple(torch.FloatTensor),optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.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 whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftorch.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-attentionheads.
TheRobertaForQuestionAnswering forward method, overrides the__call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the pre and post processing steps whilethe latter silently ignores them.
Example:
>>>from transformersimport AutoTokenizer, RobertaForQuestionAnswering>>>import torch>>>tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")>>>model = RobertaForQuestionAnswering.from_pretrained("FacebookAI/roberta-base")>>>question, text ="Who was Jim Henson?","Jim Henson was a nice puppet">>>inputs = tokenizer(question, text, return_tensors="pt")>>>with torch.no_grad():... outputs = model(**inputs)>>>answer_start_index = outputs.start_logits.argmax()>>>answer_end_index = outputs.end_logits.argmax()>>>predict_answer_tokens = inputs.input_ids[0, answer_start_index : answer_end_index +1]>>>tokenizer.decode(predict_answer_tokens, skip_special_tokens=True)...>>># target is "nice puppet">>>target_start_index = torch.tensor([14])>>>target_end_index = torch.tensor([15])>>>outputs = model(**inputs, start_positions=target_start_index, end_positions=target_end_index)>>>loss = outputs.loss>>>round(loss.item(),2)...