Transformers documentation

XLNet

Join the Hugging Face community

to get started

# XLNet

## Overview

The XLNet model was proposed in XLNet: Generalized Autoregressive Pretraining for Language Understanding by Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le. XLnet is an extension of the Transformer-XL model pre-trained using an autoregressive method to learn bidirectional contexts by maximizing the expected likelihood over all permutations of the input sequence factorization order.

The abstract from the paper is the following:

With the capability of modeling bidirectional contexts, denoising autoencoding based pretraining like BERT achieves better performance than pretraining approaches based on autoregressive language modeling. However, relying on corrupting the input with masks, BERT neglects dependency between the masked positions and suffers from a pretrain-finetune discrepancy. In light of these pros and cons, we propose XLNet, a generalized autoregressive pretraining method that (1) enables learning bidirectional contexts by maximizing the expected likelihood over all permutations of the factorization order and (2) overcomes the limitations of BERT thanks to its autoregressive formulation. Furthermore, XLNet integrates ideas from Transformer-XL, the state-of-the-art autoregressive model, into pretraining. Empirically, under comparable experiment settings, XLNet outperforms BERT on 20 tasks, often by a large margin, including question answering, natural language inference, sentiment analysis, and document ranking.

Tips:

• The specific attention pattern can be controlled at training and test time using the perm_mask input.
• Due to the difficulty of training a fully auto-regressive model over various factorization order, XLNet is pretrained using only a sub-set of the output tokens as target which are selected with the target_mapping input.
• To use XLNet for sequential decoding (i.e. not in fully bi-directional setting), use the perm_mask and target_mapping inputs to control the attention span and outputs (see examples in examples/pytorch/text-generation/run_generation.py)
• XLNet is one of the few models that has no sequence length limit.

This model was contributed by thomwolf. The original code can be found here.

## XLNetConfig

### class transformers.XLNetConfig

< >

( vocab_size = 32000 d_model = 1024 n_layer = 24 n_head = 16 d_inner = 4096 ff_activation = 'gelu' untie_r = True attn_type = 'bi' initializer_range = 0.02 layer_norm_eps = 1e-12 dropout = 0.1 mem_len = 512 reuse_len = None use_mems_eval = True use_mems_train = False bi_data = False clamp_len = -1 same_length = False summary_type = 'last' summary_use_proj = True summary_activation = 'tanh' summary_last_dropout = 0.1 start_n_top = 5 end_n_top = 5 pad_token_id = 5 bos_token_id = 1 eos_token_id = 2 **kwargs )

Parameters

• vocab_size (int, optional, defaults to 32000) — Vocabulary size of the XLNet model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling XLNetModel or TFXLNetModel.
• d_model (int, optional, defaults to 1024) — Dimensionality of the encoder layers and the pooler layer.
• n_layer (int, optional, defaults to 24) — Number of hidden layers in the Transformer encoder.
• n_head (int, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer encoder.
• d_inner (int, optional, defaults to 4096) — Dimensionality of the “intermediate” (often named feed-forward) layer in the Transformer encoder.
• ff_activation (str or Callable, optional, defaults to "gelu") — The non-linear activation function (function or string) in the If string, "gelu", "relu", "silu" and "gelu_new" are supported.
• untie_r (bool, optional, defaults to True) — Whether or not to untie relative position biases
• attn_type (str, optional, defaults to "bi") — The attention type used by the model. Set "bi" for XLNet, "uni" for Transformer-XL.
• 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.
• dropout (float, optional, defaults to 0.1) — The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
• mem_len (int or None, optional) — The number of tokens to cache. The key/value pairs that have already been pre-computed in a previous forward pass won’t be re-computed. See the quickstart for more information.
• reuse_len (int, optional) — The number of tokens in the current batch to be cached and reused in the future.
• bi_data (bool, optional, defaults to False) — Whether or not to use bidirectional input pipeline. Usually set to True during pretraining and False during finetuning.
• clamp_len (int, optional, defaults to -1) — Clamp all relative distances larger than clamp_len. Setting this attribute to -1 means no clamping.
• same_length (bool, optional, defaults to False) — Whether or not to use the same attention length for each token.
• summary_type (str, optional, defaults to “last”) — Argument used when doing sequence summary. Used in the sequence classification and multiple choice models.

Has to be one of the following options:

• "last": Take the last token hidden state (like XLNet).
• "first": Take the first token hidden state (like BERT).
• "mean": Take the mean of all tokens hidden states.
• "cls_index": Supply a Tensor of classification token position (like GPT/GPT-2).
• "attn": Not implemented now, use multi-head attention.
• summary_use_proj (bool, optional, defaults to True) — Argument used when doing sequence summary. Used in the sequence classification and multiple choice models.

Whether or not to add a projection after the vector extraction.

• summary_activation (str, optional) — Argument used when doing sequence summary. Used in the sequence classification and multiple choice models.

Pass "tanh" for a tanh activation to the output, any other value will result in no activation.

• summary_proj_to_labels (boo, optional, defaults to True) — Used in the sequence classification and multiple choice models.

Whether the projection outputs should have config.num_labels or config.hidden_size classes.

• summary_last_dropout (float, optional, defaults to 0.1) — Used in the sequence classification and multiple choice models.

The dropout ratio to be used after the projection and activation.

• start_n_top (int, optional, defaults to 5) — Used in the SQuAD evaluation script.
• end_n_top (int, optional, defaults to 5) — Used in the SQuAD evaluation script.
• use_mems_eval (bool, optional, defaults to True) — Whether or not the model should make use of the recurrent memory mechanism in evaluation mode.
• use_mems_train (bool, optional, defaults to False) — Whether or not the model should make use of the recurrent memory mechanism in train mode.

For pretraining, it is recommended to set use_mems_train to True. For fine-tuning, it is recommended to set use_mems_train to False as discussed here. If use_mems_train is set to True, one has to make sure that the train batches are correctly pre-processed, e.g. batch_1 = [[This line is], [This is the]] and batch_2 = [[ the first line], [ second line]] and that all batches are of equal size.

This is the configuration class to store the configuration of a XLNetModel or a TFXLNetModel. It is used to instantiate a XLNet 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 xlnet-large-cased architecture.

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

Examples:

>>> from transformers import XLNetConfig, XLNetModel

>>> # Initializing a XLNet configuration
>>> configuration = XLNetConfig()

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

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

## XLNetTokenizer

### class transformers.XLNetTokenizer

< >

( vocab_file do_lower_case = False remove_space = True keep_accents = False bos_token = '<s>' eos_token = '</s>' unk_token = '<unk>' sep_token = '<sep>' pad_token = '<pad>' cls_token = '<cls>' mask_token = '<mask>' additional_special_tokens = ['<eop>', '<eod>'] sp_model_kwargs: typing.Union[typing.Dict[str, typing.Any], NoneType] = None **kwargs )

Parameters

• vocab_file (str) — SentencePiece file (generally has a .spm extension) that contains the vocabulary necessary to instantiate a tokenizer.
• do_lower_case (bool, optional, defaults to True) — Whether to lowercase the input when tokenizing.
• remove_space (bool, optional, defaults to True) — Whether to strip the text when tokenizing (removing excess spaces before and after the string).
• keep_accents (bool, optional, defaults to False) — Whether to keep accents when tokenizing.
• 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 of sequence. 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.

• unk_token (str, optional, defaults to "<unk>") — The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.
• sep_token (str, optional, defaults to "<sep>") — The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens.
• pad_token (str, optional, defaults to "<pad>") — The token used for padding, for example when batching sequences of different lengths.
• cls_token (str, optional, defaults to "<cls>") — The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens.
• mask_token (str, optional, defaults to "<mask>") — The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict.
• additional_special_tokens (List[str], optional, defaults to ["<eop>", "<eod>"]) — Additional special tokens used by the tokenizer.
• sp_model_kwargs (dict, optional) — Will be passed to the SentencePieceProcessor.__init__() method. The Python wrapper for SentencePiece can be used, among other things, to set:

• enable_sampling: Enable subword regularization.

• nbest_size: Sampling parameters for unigram. Invalid for BPE-Dropout.

• nbest_size = {0,1}: No sampling is performed.
• nbest_size > 1: samples from the nbest_size results.
• nbest_size < 0: assuming that nbest_size is infinite and samples from the all hypothesis (lattice) using forward-filtering-and-backward-sampling algorithm.
• alpha: Smoothing parameter for unigram sampling, and dropout probability of merge operations for BPE-dropout.

• sp_model (SentencePieceProcessor) — The SentencePiece processor that is used for every conversion (string, tokens and IDs).

Construct an XLNet tokenizer. Based on SentencePiece.

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

#### build_inputs_with_special_tokens

< >

( token_ids_0: typing.List[int] token_ids_1: typing.Optional[typing.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 of input IDs with the appropriate special tokens.

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

• single sequence: X <sep> <cls>
• pair of sequences: A <sep> B <sep> <cls>

< >

( token_ids_0: typing.List[int] token_ids_1: typing.Optional[typing.List[int]] = None already_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 to False) — Whether or not the token list is already formatted with special tokens for the model.

Returns

List[int]

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

Retrieve 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.

#### create_token_type_ids_from_sequences

< >

( token_ids_0: typing.List[int] token_ids_1: typing.Optional[typing.List[int]] = None ) 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.

Returns

List[int]

List of token type IDs according to the given sequence(s).

Create a mask from the two sequences passed to be used in a sequence-pair classification task. An XLNet

sequence pair mask has the following format:

0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
| first sequence    | second sequence |

If token_ids_1 is None, this method only returns the first portion of the mask (0s).

#### save_vocabulary

< >

( save_directory: str filename_prefix: typing.Optional[str] = None )

## XLNetTokenizerFast

### class transformers.XLNetTokenizerFast

< >

( vocab_file = None tokenizer_file = None do_lower_case = False remove_space = True keep_accents = False bos_token = '<s>' eos_token = '</s>' unk_token = '<unk>' sep_token = '<sep>' pad_token = '<pad>' cls_token = '<cls>' mask_token = '<mask>' additional_special_tokens = ['<eop>', '<eod>'] **kwargs )

Parameters

• vocab_file (str) — SentencePiece file (generally has a .spm extension) that contains the vocabulary necessary to instantiate a tokenizer.
• do_lower_case (bool, optional, defaults to True) — Whether to lowercase the input when tokenizing.
• remove_space (bool, optional, defaults to True) — Whether to strip the text when tokenizing (removing excess spaces before and after the string).
• keep_accents (bool, optional, defaults to False) — Whether to keep accents when tokenizing.
• 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 of sequence. 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.

• unk_token (str, optional, defaults to "<unk>") — The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.
• sep_token (str, optional, defaults to "<sep>") — The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens.
• pad_token (str, optional, defaults to "<pad>") — The token used for padding, for example when batching sequences of different lengths.
• cls_token (str, optional, defaults to "<cls>") — The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens.
• mask_token (str, optional, defaults to "<mask>") — The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict.
• additional_special_tokens (List[str], optional, defaults to ["<eop>", "<eod>"]) — Additional special tokens used by the tokenizer.
• sp_model (SentencePieceProcessor) — The SentencePiece processor that is used for every conversion (string, tokens and IDs).

Construct a “fast” XLNet tokenizer (backed by HuggingFace’s tokenizers library). Based on Unigram.

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

#### build_inputs_with_special_tokens

< >

( token_ids_0: typing.List[int] token_ids_1: typing.Optional[typing.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 of input IDs with the appropriate special tokens.

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

• single sequence: X <sep> <cls>
• pair of sequences: A <sep> B <sep> <cls>

#### create_token_type_ids_from_sequences

< >

( token_ids_0: typing.List[int] token_ids_1: typing.Optional[typing.List[int]] = None ) 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.

Returns

List[int]

List of token type IDs according to the given sequence(s).

Create a mask from the two sequences passed to be used in a sequence-pair classification task. An XLNet

sequence pair mask has the following format:

0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
| first sequence    | second sequence |

If token_ids_1 is None, this method only returns the first portion of the mask (0s).

## XLNet specific outputs

### class transformers.models.xlnet.modeling_xlnet.XLNetModelOutput

< >

( last_hidden_state: FloatTensor mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• last_hidden_state (torch.FloatTensor of shape (batch_size, num_predict, hidden_size)) — Sequence of hidden-states at the last layer of the model.

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of XLNetModel.

< >

( loss: typing.Optional[torch.FloatTensor] = None logits: FloatTensor = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).
• logits (torch.FloatTensor of shape (batch_size, num_predict, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

### class transformers.models.xlnet.modeling_xlnet.XLNetForSequenceClassificationOutput

< >

( loss: typing.Optional[torch.FloatTensor] = None logits: FloatTensor = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• 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).
• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of XLNetForSequenceClassification.

### class transformers.models.xlnet.modeling_xlnet.XLNetForMultipleChoiceOutput

< >

( loss: typing.Optional[torch.FloatTensor] = None logits: FloatTensor = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification loss.
• logits (torch.FloatTensor of shape (batch_size, num_choices)) — num_choices is the second dimension of the input tensors. (see input_ids above).

Classification scores (before SoftMax).

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of XLNetForMultipleChoice.

### class transformers.models.xlnet.modeling_xlnet.XLNetForTokenClassificationOutput

< >

( loss: typing.Optional[torch.FloatTensor] = None logits: FloatTensor = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification loss.
• logits (torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels)) — Classification scores (before SoftMax).
• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of XLNetForTokenClassificationOutput.

< >

( loss: typing.Optional[torch.FloatTensor] = None start_logits: FloatTensor = None end_logits: FloatTensor = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.
• start_logits (torch.FloatTensor of shape (batch_size, sequence_length,)) — Span-start scores (before SoftMax).
• end_logits (torch.FloatTensor of shape (batch_size, sequence_length,)) — Span-end scores (before SoftMax).
• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

< >

( loss: typing.Optional[torch.FloatTensor] = None start_top_log_probs: typing.Optional[torch.FloatTensor] = None start_top_index: typing.Optional[torch.LongTensor] = None end_top_log_probs: typing.Optional[torch.FloatTensor] = None end_top_index: typing.Optional[torch.LongTensor] = None cls_logits: typing.Optional[torch.FloatTensor] = None mems: typing.Optional[typing.List[torch.FloatTensor]] = None hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

• loss (torch.FloatTensor of shape (1,), optional, returned if both start_positions and end_positions are provided) — Classification loss as the sum of start token, end token (and is_impossible if provided) classification losses.
• start_top_log_probs (torch.FloatTensor of shape (batch_size, config.start_n_top), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the top config.start_n_top start token possibilities (beam-search).
• start_top_index (torch.LongTensor of shape (batch_size, config.start_n_top), optional, returned if start_positions or end_positions is not provided) — Indices for the top config.start_n_top start token possibilities (beam-search).
• end_top_log_probs (torch.FloatTensor of shape (batch_size, config.start_n_top * config.end_n_top), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the top config.start_n_top * config.end_n_top end token possibilities (beam-search).
• end_top_index (torch.LongTensor of shape (batch_size, config.start_n_top * config.end_n_top), optional, returned if start_positions or end_positions is not provided) — Indices for the top config.start_n_top * config.end_n_top end token possibilities (beam-search).
• cls_logits (torch.FloatTensor of shape (batch_size,), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the is_impossible label of the answers.
• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

### class transformers.models.xlnet.modeling_tf_xlnet.TFXLNetModelOutput

< >

( last_hidden_state: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• last_hidden_state (tf.Tensor of shape (batch_size, num_predict, hidden_size)) — Sequence of hidden-states at the last layer of the model.

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of TFXLNetModel.

< >

( loss: typing.Optional[tensorflow.python.framework.ops.Tensor] = None logits: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).
• logits (tf.Tensor of shape (batch_size, num_predict, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

### class transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForSequenceClassificationOutput

< >

( loss: typing.Optional[tensorflow.python.framework.ops.Tensor] = None logits: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• loss (tf.Tensor of shape (1,), optional, returned when label is provided) — Classification (or regression if config.num_labels==1) loss.
• logits (tf.Tensor of shape (batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax).
• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of TFXLNetForSequenceClassification.

### class transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForMultipleChoiceOutput

< >

( loss: typing.Optional[tensorflow.python.framework.ops.Tensor] = None logits: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Classification loss.
• logits (tf.Tensor of shape (batch_size, num_choices)) — num_choices is the second dimension of the input tensors. (see input_ids above).

Classification scores (before SoftMax).

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of TFXLNetForMultipleChoice.

### class transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForTokenClassificationOutput

< >

( loss: typing.Optional[tensorflow.python.framework.ops.Tensor] = None logits: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Classification loss.
• logits (tf.Tensor of shape (batch_size, sequence_length, config.num_labels)) — Classification scores (before SoftMax).
• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

Output type of TFXLNetForTokenClassificationOutput.

< >

( loss: typing.Optional[tensorflow.python.framework.ops.Tensor] = None start_logits: Tensor = None end_logits: Tensor = None mems: typing.Optional[typing.List[tensorflow.python.framework.ops.Tensor]] = None hidden_states: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None attentions: typing.Optional[typing.Tuple[tensorflow.python.framework.ops.Tensor]] = None )

Parameters

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.
• start_logits (tf.Tensor of shape (batch_size, sequence_length,)) — Span-start scores (before SoftMax).
• end_logits (tf.Tensor of shape (batch_size, sequence_length,)) — Span-end scores (before SoftMax).
• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.
• 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.

## XLNetModel

### class transformers.XLNetModel

< >

( config )

Parameters

• config (XLNetConfig) — 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.

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

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetModelOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.xlnet.modeling_xlnet.XLNetModelOutput or tuple(torch.FloatTensor)

A transformers.models.xlnet.modeling_xlnet.XLNetModelOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• last_hidden_state (torch.FloatTensor of shape (batch_size, num_predict, hidden_size)) — Sequence of hidden-states at the last layer of the model.

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetModel forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, XLNetModel
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = XLNetModel.from_pretrained("xlnet-base-cased")

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

>>> last_hidden_states = outputs.last_hidden_state

< >

( config )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a language modeling head on top (linear layer with weights tied to the input embeddings).

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetLMHeadModelOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (torch.LongTensor of shape (batch_size, num_predict), optional) — Labels for masked language modeling. num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

The labels should correspond to the masked input words that should be predicted and depends on target_mapping. Note in order to perform standard auto-regressive language modeling a token has to be added to the input_ids (see the prepare_inputs_for_generation function and examples below)

Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored, the loss is only computed for labels in [0, ..., config.vocab_size]

Returns

transformers.models.xlnet.modeling_xlnet.XLNetLMHeadModelOutput or tuple(torch.FloatTensor)

A transformers.models.xlnet.modeling_xlnet.XLNetLMHeadModelOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) Language modeling loss (for next-token prediction).

• logits (torch.FloatTensor of shape (batch_size, num_predict, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetLMHeadModel forward method, overrides the __call__ special method.

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

Examples:

>>> from transformers import XLNetTokenizer, XLNetLMHeadModel
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-large-cased")

>>> # We show how to setup inputs to predict a next token using a bi-directional context.
>>> input_ids = torch.tensor(
... ).unsqueeze(
...     0
... )  # We will predict the masked token
>>> perm_mask = torch.zeros((1, input_ids.shape[1], input_ids.shape[1]), dtype=torch.float)
>>> perm_mask[:, :, -1] = 1.0  # Previous tokens don't see last token
>>> target_mapping = torch.zeros(
...     (1, 1, input_ids.shape[1]), dtype=torch.float
... )  # Shape [1, 1, seq_length] => let's predict one token
>>> target_mapping[
...     0, 0, -1
... ] = 1.0  # Our first (and only) prediction will be the last token of the sequence (the masked token)

>>> next_token_logits = outputs[
...     0
... ]  # Output has shape [target_mapping.size(0), target_mapping.size(1), config.vocab_size]

>>> # The same way can the XLNetLMHeadModel be used to be trained by standard auto-regressive language modeling.
>>> input_ids = torch.tensor(
... ).unsqueeze(
...     0
... )  # We will predict the masked token
>>> assert labels.shape[0] == 1, "only one word will be predicted"
>>> perm_mask = torch.zeros((1, input_ids.shape[1], input_ids.shape[1]), dtype=torch.float)
...     :, :, -1
... ] = 1.0  # Previous tokens don't see last token as is done in standard auto-regressive lm training
>>> target_mapping = torch.zeros(
...     (1, 1, input_ids.shape[1]), dtype=torch.float
... )  # Shape [1, 1, seq_length] => let's predict one token
>>> target_mapping[
...     0, 0, -1
... ] = 1.0  # Our first (and only) prediction will be the last token of the sequence (the masked token)

>>> loss = outputs.loss
>>> next_token_logits = (
...     outputs.logits
... )  # Logits have shape [target_mapping.size(0), target_mapping.size(1), config.vocab_size]

## XLNetForSequenceClassification

### class transformers.XLNetForSequenceClassification

< >

( config )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetForSequenceClassificationOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (torch.LongTensor of shape (batch_size,), optional) — Labels for computing the sequence classification/regression loss. Indices should be in [0, ..., config.num_labels - 1]. If config.num_labels == 1 a regression loss is computed (Mean-Square loss), If config.num_labels > 1 a classification loss is computed (Cross-Entropy).

A transformers.models.xlnet.modeling_xlnet.XLNetForSequenceClassificationOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• 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).

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetForSequenceClassification forward method, overrides the __call__ special method.

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

Example of single-label classification:

>>> import torch
>>> from transformers import XLNetTokenizer, XLNetForSequenceClassification

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = XLNetForSequenceClassification.from_pretrained("xlnet-base-cased")

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

...     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 = XLNetForSequenceClassification.from_pretrained("xlnet-base-cased", 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 transformers import XLNetTokenizer, XLNetForSequenceClassification

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = XLNetForSequenceClassification.from_pretrained("xlnet-base-cased", problem_type="multi_label_classification")

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

...     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 = XLNetForSequenceClassification.from_pretrained(
...     "xlnet-base-cased", num_labels=num_labels, problem_type="multi_label_classification"
... )

>>> labels = torch.nn.functional.one_hot(torch.tensor([predicted_class_id]), num_classes=num_labels).to(
...     torch.float
... )
>>> loss = model(**inputs, labels=labels).loss
>>> loss.backward()

## XLNetForMultipleChoice

### class transformers.XLNetForMultipleChoice

< >

( config )

Parameters

• config (XLNetConfig) — 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.

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

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetForMultipleChoiceOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, num_choices, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, num_choices, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, num_choices, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, num_choices, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, num_choices, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (torch.LongTensor of shape (batch_size,), optional) — Labels for computing the multiple choice classification loss. Indices should be in [0, ..., num_choices-1] where num_choices is the size of the second dimension of the input tensors. (See input_ids above)

Returns

transformers.models.xlnet.modeling_xlnet.XLNetForMultipleChoiceOutput or tuple(torch.FloatTensor)

A transformers.models.xlnet.modeling_xlnet.XLNetForMultipleChoiceOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

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

• logits (torch.FloatTensor of shape (batch_size, num_choices)) — num_choices is the second dimension of the input tensors. (see input_ids above).

Classification scores (before SoftMax).

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetForMultipleChoice forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, XLNetForMultipleChoice
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = XLNetForMultipleChoice.from_pretrained("xlnet-base-cased")

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

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

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

## XLNetForTokenClassification

### class transformers.XLNetForTokenClassification

< >

( config )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None labels: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetForTokenClassificationOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (torch.LongTensor of shape (batch_size,), optional) — Labels for computing the multiple choice classification loss. Indices should be in [0, ..., num_choices] where num_choices is the size of the second dimension of the input tensors. (see input_ids above)

Returns

transformers.models.xlnet.modeling_xlnet.XLNetForTokenClassificationOutput or tuple(torch.FloatTensor)

A transformers.models.xlnet.modeling_xlnet.XLNetForTokenClassificationOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

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

• logits (torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels)) — Classification scores (before SoftMax).

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetForTokenClassification forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, XLNetForTokenClassification
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = XLNetForTokenClassification.from_pretrained("xlnet-base-cased")

>>> inputs = tokenizer(
...     "HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="pt"
... )

...     logits = model(**inputs).logits

>>> predicted_token_class_ids = logits.argmax(-1)

>>> # Note that tokens are classified rather then input words which means that
>>> # there might be more predicted token classes than words.
>>> # Multiple token classes might account for the same word
>>> predicted_tokens_classes = [model.config.id2label[t.item()] for t in predicted_token_class_ids[0]]
>>> predicted_tokens_classes

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


< >

( config )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of the hidden-states output to compute span start logits and span end logits).

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None start_positions: typing.Optional[torch.Tensor] = None end_positions: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetForQuestionAnsweringSimpleOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• start_positions (torch.LongTensor of shape (batch_size,), optional) — Labels for position (index) of the start of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length). Position outside of the sequence are not taken into account for computing the loss.
• end_positions (torch.LongTensor of shape (batch_size,), optional) — Labels for position (index) of the end of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length). Position outside of the sequence are not taken into account for computing the loss.

A transformers.models.xlnet.modeling_xlnet.XLNetForQuestionAnsweringSimpleOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Total span extraction loss is the sum of a Cross-Entropy for the start and end positions.

• start_logits (torch.FloatTensor of shape (batch_size, sequence_length,)) — Span-start scores (before SoftMax).

• end_logits (torch.FloatTensor of shape (batch_size, sequence_length,)) — Span-end scores (before SoftMax).

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetForQuestionAnsweringSimple forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, XLNetForQuestionAnsweringSimple
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")

>>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"

>>> inputs = tokenizer(question, text, return_tensors="pt")
...     outputs = model(**inputs)


>>> # 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)


< >

( config )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of the hidden-states output to compute span start logits and span end logits).

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

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

#### forward

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None mems: typing.Optional[torch.Tensor] = None perm_mask: typing.Optional[torch.Tensor] = None target_mapping: typing.Optional[torch.Tensor] = None token_type_ids: typing.Optional[torch.Tensor] = None input_mask: typing.Optional[torch.Tensor] = None head_mask: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None start_positions: typing.Optional[torch.Tensor] = None end_positions: typing.Optional[torch.Tensor] = None is_impossible: typing.Optional[torch.Tensor] = None cls_index: typing.Optional[torch.Tensor] = None p_mask: typing.Optional[torch.Tensor] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) transformers.models.xlnet.modeling_xlnet.XLNetForQuestionAnsweringOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• start_positions (torch.LongTensor of shape (batch_size,), optional) — Labels for position (index) of the start of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length). Position outside of the sequence are not taken into account for computing the loss.
• end_positions (torch.LongTensor of shape (batch_size,), optional) — Labels for position (index) of the end of the labelled span for computing the token classification loss. Positions are clamped to the length of the sequence (sequence_length). Position outside of the sequence are not taken into account for computing the loss.
• is_impossible (torch.LongTensor of shape (batch_size,), optional) — Labels whether a question has an answer or no answer (SQuAD 2.0)
• cls_index (torch.LongTensor of shape (batch_size,), optional) — Labels for position (index) of the classification token to use as input for computing plausibility of the answer.
• p_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Optional mask of tokens which can’t be in answers (e.g. [CLS], [PAD], …). 1.0 means token should be masked. 0.0 mean token is not masked.

Returns

transformers.models.xlnet.modeling_xlnet.XLNetForQuestionAnsweringOutput or tuple(torch.FloatTensor)

A transformers.models.xlnet.modeling_xlnet.XLNetForQuestionAnsweringOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned if both start_positions and end_positions are provided) — Classification loss as the sum of start token, end token (and is_impossible if provided) classification losses.

• start_top_log_probs (torch.FloatTensor of shape (batch_size, config.start_n_top), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the top config.start_n_top start token possibilities (beam-search).

• start_top_index (torch.LongTensor of shape (batch_size, config.start_n_top), optional, returned if start_positions or end_positions is not provided) — Indices for the top config.start_n_top start token possibilities (beam-search).

• end_top_log_probs (torch.FloatTensor of shape (batch_size, config.start_n_top * config.end_n_top), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the top config.start_n_top * config.end_n_top end token possibilities (beam-search).

• end_top_index (torch.LongTensor of shape (batch_size, config.start_n_top * config.end_n_top), optional, returned if start_positions or end_positions is not provided) — Indices for the top config.start_n_top * config.end_n_top end token possibilities (beam-search).

• cls_logits (torch.FloatTensor of shape (batch_size,), optional, returned if start_positions or end_positions is not provided) — Log probabilities for the is_impossible label of the answers.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• 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.

The XLNetForQuestionAnswering forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, XLNetForQuestionAnswering
>>> import torch

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")

>>> input_ids = torch.tensor(tokenizer.encode("Hello, my dog is cute", add_special_tokens=True)).unsqueeze(
...     0
... )  # Batch size 1
>>> start_positions = torch.tensor([1])
>>> end_positions = torch.tensor([3])
>>> outputs = model(input_ids, start_positions=start_positions, end_positions=end_positions)

>>> loss = outputs.loss

## TFXLNetModel

### class transformers.TFXLNetModel

< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

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

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})

#### call

< >

( input_ids = None attention_mask = None mems = None perm_mask = None target_mapping = None token_type_ids = None input_mask = None head_mask = None inputs_embeds = None use_mems = None output_attentions = None output_hidden_states = None return_dict = None training = False ) transformers.models.xlnet.modeling_tf_xlnet.TFXLNetModelOutput or tuple(tf.Tensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.xlnet.modeling_tf_xlnet.TFXLNetModelOutput or tuple(tf.Tensor)

A transformers.models.xlnet.modeling_tf_xlnet.TFXLNetModelOutput or a tuple of tf.Tensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• last_hidden_state (tf.Tensor of shape (batch_size, num_predict, hidden_size)) — Sequence of hidden-states at the last layer of the model.

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

• attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The TFXLNetModel forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, TFXLNetModel
>>> import tensorflow as tf

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = TFXLNetModel.from_pretrained("xlnet-base-cased")

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

>>> last_hidden_states = outputs.last_hidden_state

< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a language modeling head on top (linear layer with weights tied to the input embeddings).

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})

#### call

< >

( input_ids: typing.Union[typing.List[tensorflow.python.framework.ops.Tensor], typing.List[numpy.ndarray], typing.List[tensorflow.python.keras.engine.keras_tensor.KerasTensor], typing.Dict[str, tensorflow.python.framework.ops.Tensor], typing.Dict[str, numpy.ndarray], typing.Dict[str, tensorflow.python.keras.engine.keras_tensor.KerasTensor], tensorflow.python.framework.ops.Tensor, numpy.ndarray, tensorflow.python.keras.engine.keras_tensor.KerasTensor, NoneType] = None attention_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None mems: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None perm_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None target_mapping: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None token_type_ids: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None input_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None head_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None inputs_embeds: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None labels: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None training: typing.Optional[bool] = False ) transformers.models.xlnet.modeling_tf_xlnet.TFXLNetLMHeadModelOutput or tuple(tf.Tensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (tf.Tensor of shape (batch_size, sequence_length), optional) — Labels for computing the cross entropy classification loss. Indices should be in [0, ..., config.vocab_size - 1].

A transformers.models.xlnet.modeling_tf_xlnet.TFXLNetLMHeadModelOutput or a tuple of tf.Tensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) Language modeling loss (for next-token prediction).

• logits (tf.Tensor of shape (batch_size, num_predict, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

num_predict corresponds to target_mapping.shape[1]. If target_mapping is None, then num_predict corresponds to sequence_length.

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

• attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The TFXLNetLMHeadModel forward method, overrides the __call__ special method.

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

Examples:

>>> import tensorflow as tf
>>> import numpy as np
>>> from transformers import XLNetTokenizer, TFXLNetLMHeadModel

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-large-cased")

>>> # We show how to setup inputs to predict a next token using a bi-directional context.
...     None, :
... ]  # We will predict the masked token

>>> perm_mask = np.zeros((1, input_ids.shape[1], input_ids.shape[1]))
>>> perm_mask[:, :, -1] = 1.0  # Previous tokens don't see last token

>>> target_mapping = np.zeros(
...     (1, 1, input_ids.shape[1])
... )  # Shape [1, 1, seq_length] => let's predict one token
>>> target_mapping[
...     0, 0, -1
... ] = 1.0  # Our first (and only) prediction will be the last token of the sequence (the masked token)

>>> outputs = model(
...     input_ids,
...     target_mapping=tf.constant(target_mapping, dtype=tf.float32),
... )

>>> next_token_logits = outputs[
...     0
... ]  # Output has shape [target_mapping.size(0), target_mapping.size(1), config.vocab_size]

## TFXLNetForSequenceClassification

### class transformers.TFXLNetForSequenceClassification

< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks.

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})

#### call

< >

( input_ids: typing.Union[typing.List[tensorflow.python.framework.ops.Tensor], typing.List[numpy.ndarray], typing.List[tensorflow.python.keras.engine.keras_tensor.KerasTensor], typing.Dict[str, tensorflow.python.framework.ops.Tensor], typing.Dict[str, numpy.ndarray], typing.Dict[str, tensorflow.python.keras.engine.keras_tensor.KerasTensor], tensorflow.python.framework.ops.Tensor, numpy.ndarray, tensorflow.python.keras.engine.keras_tensor.KerasTensor, NoneType] = None attention_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None mems: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None perm_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None target_mapping: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None token_type_ids: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None input_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None head_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None inputs_embeds: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None labels: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None training: typing.Optional[bool] = False ) transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForSequenceClassificationOutput or tuple(tf.Tensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (tf.Tensor of shape (batch_size,), optional) — Labels for computing the sequence classification/regression loss. Indices should be in [0, ..., config.num_labels - 1]. If config.num_labels == 1 a regression loss is computed (Mean-Square loss), If config.num_labels > 1 a classification loss is computed (Cross-Entropy).

A transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForSequenceClassificationOutput or a tuple of tf.Tensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (tf.Tensor of shape (1,), optional, returned when label is provided) — Classification (or regression if config.num_labels==1) loss.

• logits (tf.Tensor of shape (batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax).

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

• attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The TFXLNetForSequenceClassification forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, TFXLNetForSequenceClassification
>>> import tensorflow as tf

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = TFXLNetForSequenceClassification.from_pretrained("xlnet-base-cased")

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

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

>>> predicted_class_id = int(tf.math.argmax(logits, axis=-1)[0])
>>> model.config.id2label[predicted_class_id]

>>> # 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 = TFXLNetForSequenceClassification.from_pretrained("xlnet-base-cased", num_labels=num_labels)

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


## TFLNetForMultipleChoice

### class transformers.TFXLNetForMultipleChoice

< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

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

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})

#### call

< >

( input_ids: typing.Union[typing.List[tensorflow.python.framework.ops.Tensor], typing.List[numpy.ndarray], typing.List[tensorflow.python.keras.engine.keras_tensor.KerasTensor], typing.Dict[str, tensorflow.python.framework.ops.Tensor], typing.Dict[str, numpy.ndarray], typing.Dict[str, tensorflow.python.keras.engine.keras_tensor.KerasTensor], tensorflow.python.framework.ops.Tensor, numpy.ndarray, tensorflow.python.keras.engine.keras_tensor.KerasTensor, NoneType] = None token_type_ids: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None input_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None attention_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None mems: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None perm_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None target_mapping: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None head_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None inputs_embeds: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None labels: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None training: typing.Optional[bool] = False ) transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForMultipleChoiceOutput or tuple(tf.Tensor)

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 XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, num_choices, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, num_choices, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, num_choices, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, num_choices, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (tf.Tensor of shape (batch_size,), optional) — Labels for computing the multiple choice classification loss. Indices should be in [0, ..., num_choices] where num_choices is the size of the second dimension of the input tensors. (See input_ids above)

A transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForMultipleChoiceOutput or a tuple of tf.Tensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Classification loss.

• logits (tf.Tensor of shape (batch_size, num_choices)) — num_choices is the second dimension of the input tensors. (see input_ids above).

Classification scores (before SoftMax).

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

• attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The TFXLNetForMultipleChoice forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, TFXLNetForMultipleChoice
>>> import tensorflow as tf

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = TFXLNetForMultipleChoice.from_pretrained("xlnet-base-cased")

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

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

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

## TFXLNetForTokenClassification

### class transformers.TFXLNetForTokenClassification

< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks.

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})

#### call

< >

( input_ids: typing.Union[typing.List[tensorflow.python.framework.ops.Tensor], typing.List[numpy.ndarray], typing.List[tensorflow.python.keras.engine.keras_tensor.KerasTensor], typing.Dict[str, tensorflow.python.framework.ops.Tensor], typing.Dict[str, numpy.ndarray], typing.Dict[str, tensorflow.python.keras.engine.keras_tensor.KerasTensor], tensorflow.python.framework.ops.Tensor, numpy.ndarray, tensorflow.python.keras.engine.keras_tensor.KerasTensor, NoneType] = None attention_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None mems: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None perm_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None target_mapping: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None token_type_ids: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None input_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None head_mask: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None inputs_embeds: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None use_mems: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None labels: typing.Union[numpy.ndarray, tensorflow.python.framework.ops.Tensor, NoneType] = None training: typing.Optional[bool] = False ) transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForTokenClassificationOutput or tuple(tf.Tensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary.

Indices can be obtained using XLNetTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• mems (List[torch.FloatTensor] of length config.n_layers) — Contains pre-computed hidden-states (see mems output below) . Can be used to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

use_mems has to be set to True to make use of mems.

• perm_mask (torch.FloatTensor of shape (batch_size, sequence_length, sequence_length), optional) — Mask to indicate the attention pattern for each input token with values selected in [0, 1]:

• if perm_mask[k, i, j] = 0, i attend to j in batch k;
• if perm_mask[k, i, j] = 1, i does not attend to j in batch k.

If not set, each token attends to all the others (full bidirectional attention). Only used during pretraining (to define factorization order) or for sequential decoding (generation).

• target_mapping (torch.FloatTensor of shape (batch_size, num_predict, sequence_length), optional) — Mask to indicate the output tokens to use. If target_mapping[k, i, j] = 1, the i-th predict in batch k is on the j-th token. Only used during pretraining for partial prediction or for sequential decoding (generation).
• token_type_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

• 0 corresponds to a sentence A token,
• 1 corresponds to a sentence B token.

What are token type IDs?

• input_mask (torch.FloatTensor of shape batch_size, sequence_length, optional) — Mask to avoid performing attention on padding token indices. Negative of attention_mask, i.e. with 0 for real tokens and 1 for padding which is kept for compatibility with the original code base.

Mask values selected in [0, 1]:

• 1 for tokens that are masked,
• 0 for tokens that are not masked.

You can only uses one of input_mask and attention_mask.

• head_mask (torch.FloatTensor of shape (num_heads,) or (num_layers, num_heads), optional) — Mask to nullify selected heads of the self-attention modules. Mask values selected in [0, 1]:

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
• labels (tf.Tensor of shape (batch_size, sequence_length), optional) — Labels for computing the token classification loss. Indices should be in [0, ..., config.num_labels - 1].

A transformers.models.xlnet.modeling_tf_xlnet.TFXLNetForTokenClassificationOutput or a tuple of tf.Tensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (XLNetConfig) and inputs.

• loss (tf.Tensor of shape (1,), optional, returned when labels is provided) — Classification loss.

• logits (tf.Tensor of shape (batch_size, sequence_length, config.num_labels)) — Classification scores (before SoftMax).

• mems (List[tf.Tensor] of length config.n_layers) — Contains pre-computed hidden-states. Can be used (see mems input) to speed up sequential decoding. The token ids which have their past given to this model should not be passed as input_ids as they have already been computed.

• hidden_states (tuple(tf.Tensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of tf.Tensor (one for the output of the embeddings + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the initial embedding outputs.

• attentions (tuple(tf.Tensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of tf.Tensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The TFXLNetForTokenClassification forward method, overrides the __call__ special method.

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

Example:

>>> from transformers import XLNetTokenizer, TFXLNetForTokenClassification
>>> import tensorflow as tf

>>> tokenizer = XLNetTokenizer.from_pretrained("xlnet-base-cased")
>>> model = TFXLNetForTokenClassification.from_pretrained("xlnet-base-cased")

>>> inputs = tokenizer(
...     "HuggingFace is a company based in Paris and New York", add_special_tokens=False, return_tensors="tf"
... )

>>> logits = model(**inputs).logits
>>> predicted_token_class_ids = tf.math.argmax(logits, axis=-1)

>>> # Note that tokens are classified rather then input words which means that
>>> # there might be more predicted token classes than words.
>>> # Multiple token classes might account for the same word
>>> predicted_tokens_classes = [model.config.id2label[t] for t in predicted_token_class_ids[0].numpy().tolist()]
>>> predicted_tokens_classes

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


< >

( *args **kwargs )

Parameters

• config (XLNetConfig) — 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.

XLNet Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear layers on top of the hidden-states output to compute span start logits and span end logits).

This model inherits from TFPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a tf.keras.Model subclass. Use it as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and behavior.

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})