Hugging Face's logo

Transformers documentation

DPR

You are viewing v4.27.0 version. A newer version v4.46.3 is available.
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

DPR

Models Spaces

Overview

Dense Passage Retrieval (DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. It was introduced in Dense Passage Retrieval for Open-Domain Question Answering by Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-tau Yih.

The abstract from the paper is the following:

Open-domain question answering relies on efficient passage retrieval to select candidate contexts, where traditional sparse vector space models, such as TF-IDF or BM25, are the de facto method. In this work, we show that retrieval can be practically implemented using dense representations alone, where embeddings are learned from a small number of questions and passages by a simple dual-encoder framework. When evaluated on a wide range of open-domain QA datasets, our dense retriever outperforms a strong Lucene-BM25 system largely by 9%-19% absolute in terms of top-20 passage retrieval accuracy, and helps our end-to-end QA system establish new state-of-the-art on multiple open-domain QA benchmarks.

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

Tips:

  • DPR consists in three models:

    • Question encoder: encode questions as vectors
    • Context encoder: encode contexts as vectors
    • Reader: extract the answer of the questions inside retrieved contexts, along with a relevance score (high if the inferred span actually answers the question).

DPRConfig

class transformers.DPRConfig

< >

( vocab_size = 30522 hidden_size = 768 num_hidden_layers = 12 num_attention_heads = 12 intermediate_size = 3072 hidden_act = 'gelu' hidden_dropout_prob = 0.1 attention_probs_dropout_prob = 0.1 max_position_embeddings = 512 type_vocab_size = 2 initializer_range = 0.02 layer_norm_eps = 1e-12 pad_token_id = 0 position_embedding_type = 'absolute' projection_dim: int = 0 **kwargs )

Parameters

  • vocab_size (int, optional, defaults to 30522) — Vocabulary size of the DPR model. Defines the different tokens that can be represented by the inputs_ids passed to the forward method of BertModel.
  • hidden_size (int, optional, defaults to 768) — Dimensionality of the encoder layers and the pooler layer.
  • num_hidden_layers (int, optional, defaults to 12) — Number of hidden layers in the Transformer encoder.
  • num_attention_heads (int, optional, defaults to 12) — Number of attention heads for each attention layer in the Transformer encoder.
  • intermediate_size (int, optional, defaults to 3072) — Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder.
  • hidden_act (str or function, optional, defaults to "gelu") — The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu", "relu", "silu" and "gelu_new" are supported.
  • hidden_dropout_prob (float, optional, defaults to 0.1) — The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
  • attention_probs_dropout_prob (float, optional, defaults to 0.1) — The dropout ratio for the attention probabilities.
  • max_position_embeddings (int, optional, defaults to 512) — The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048).
  • type_vocab_size (int, optional, defaults to 2) — The vocabulary size of the token_type_ids passed into BertModel.
  • 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.
  • position_embedding_type (str, optional, defaults to "absolute") — Type of position embedding. Choose one of "absolute", "relative_key", "relative_key_query". For positional embeddings use "absolute". For more information on "relative_key", please refer to Self-Attention with Relative Position Representations (Shaw et al.). For more information on "relative_key_query", please refer to Method 4 in Improve Transformer Models with Better Relative Position Embeddings (Huang et al.).
  • projection_dim (int, optional, defaults to 0) — Dimension of the projection for the context and question encoders. If it is set to zero (default), then no projection is done.

DPRConfig is the configuration class to store the configuration of a DPRModel.

This is the configuration class to store the configuration of a DPRContextEncoder, DPRQuestionEncoder, or a DPRReader. It is used to instantiate the components of the DPR model according to the specified arguments, defining the model component architectures. Instantiating a configuration with the defaults will yield a similar configuration to that of the DPRContextEncoder facebook/dpr-ctx_encoder-single-nq-base architecture.

This class is a subclass of BertConfig. Please check the superclass for the documentation of all kwargs.

Example:

>>> from transformers import DPRConfig, DPRContextEncoder

>>> # Initializing a DPR facebook/dpr-ctx_encoder-single-nq-base style configuration
>>> configuration = DPRConfig()

>>> # Initializing a model (with random weights) from the facebook/dpr-ctx_encoder-single-nq-base style configuration
>>> model = DPRContextEncoder(configuration)

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

DPRContextEncoderTokenizer

class transformers.DPRContextEncoderTokenizer

< >

( vocab_file do_lower_case = True do_basic_tokenize = True never_split = None unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs )

Construct a DPRContextEncoder tokenizer.

DPRContextEncoderTokenizer is identical to BertTokenizer and runs end-to-end tokenization: punctuation splitting and wordpiece.

Refer to superclass BertTokenizer for usage examples and documentation concerning parameters.

DPRContextEncoderTokenizerFast

class transformers.DPRContextEncoderTokenizerFast

< >

( vocab_file = None tokenizer_file = None do_lower_case = True unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs )

Construct a “fast” DPRContextEncoder tokenizer (backed by HuggingFace’s tokenizers library).

DPRContextEncoderTokenizerFast is identical to BertTokenizerFast and runs end-to-end tokenization: punctuation splitting and wordpiece.

Refer to superclass BertTokenizerFast for usage examples and documentation concerning parameters.

DPRQuestionEncoderTokenizer

class transformers.DPRQuestionEncoderTokenizer

< >

( vocab_file do_lower_case = True do_basic_tokenize = True never_split = None unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs )

Constructs a DPRQuestionEncoder tokenizer.

DPRQuestionEncoderTokenizer is identical to BertTokenizer and runs end-to-end tokenization: punctuation splitting and wordpiece.

Refer to superclass BertTokenizer for usage examples and documentation concerning parameters.

DPRQuestionEncoderTokenizerFast

class transformers.DPRQuestionEncoderTokenizerFast

< >

( vocab_file = None tokenizer_file = None do_lower_case = True unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs )

Constructs a “fast” DPRQuestionEncoder tokenizer (backed by HuggingFace’s tokenizers library).

DPRQuestionEncoderTokenizerFast is identical to BertTokenizerFast and runs end-to-end tokenization: punctuation splitting and wordpiece.

Refer to superclass BertTokenizerFast for usage examples and documentation concerning parameters.

DPRReaderTokenizer

class transformers.DPRReaderTokenizer

< >

( vocab_file do_lower_case = True do_basic_tokenize = True never_split = None unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs ) → Dict[str, List[List[int]]]

Parameters

  • questions (str or List[str]) — The questions to be encoded. You can specify one question for many passages. In this case, the question will be duplicated like [questions] * n_passages. Otherwise you have to specify as many questions as in titles or texts.
  • titles (str or List[str]) — The passages titles to be encoded. This can be a string or a list of strings if there are several passages.
  • texts (str or List[str]) — The passages texts to be encoded. This can be a string or a list of strings if there are several passages.
  • padding (bool, str or PaddingStrategy, optional, defaults to False) — Activates and controls padding. Accepts the following values:

    • True or 'longest': Pad to the longest sequence in the batch (or no padding if only a single sequence if provided).
    • 'max_length': Pad to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided.
    • False or 'do_not_pad' (default): No padding (i.e., can output a batch with sequences of different lengths).
  • truncation (bool, str or TruncationStrategy, optional, defaults to False) — Activates and controls truncation. Accepts the following values:

    • True or 'longest_first': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will truncate token by token, removing a token from the longest sequence in the pair if a pair of sequences (or a batch of pairs) is provided.
    • 'only_first': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the first sequence of a pair if a pair of sequences (or a batch of pairs) is provided.
    • 'only_second': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the second sequence of a pair if a pair of sequences (or a batch of pairs) is provided.
    • False or 'do_not_truncate' (default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size).
  • max_length (int, optional) — Controls the maximum length to use by one of the truncation/padding parameters.

    If left unset or set to None, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated.

  • return_tensors (str or TensorType, optional) — If set, will return tensors instead of list of python integers. Acceptable values are:

    • 'tf': Return TensorFlow tf.constant objects.
    • 'pt': Return PyTorch torch.Tensor objects.
    • 'np': Return Numpy np.ndarray objects.
  • return_attention_mask (bool, optional) — Whether or not to return the attention mask. If not set, will return the attention mask according to the specific tokenizer’s default, defined by the return_outputs attribute.

    What are attention masks?

Returns

Dict[str, List[List[int]]]

A dictionary with the following keys:

  • input_ids: List of token ids to be fed to a model.
  • attention_mask: List of indices specifying which tokens should be attended to by the model.

Construct a DPRReader tokenizer.

DPRReaderTokenizer is almost identical to BertTokenizer and runs end-to-end tokenization: punctuation splitting and wordpiece. The difference is that is has three inputs strings: question, titles and texts that are combined to be fed to the DPRReader model.

Refer to superclass BertTokenizer for usage examples and documentation concerning parameters.

Return a dictionary with the token ids of the input strings and other information to give to .decode_best_spans. It converts the strings of a question and different passages (title and text) in a sequence of IDs (integers), using the tokenizer and vocabulary. The resulting input_ids is a matrix of size (n_passages, sequence_length)

with the format:

[CLS] <question token ids> [SEP] <titles ids> [SEP] <texts ids>

DPRReaderTokenizerFast

class transformers.DPRReaderTokenizerFast

< >

( vocab_file = None tokenizer_file = None do_lower_case = True unk_token = '[UNK]' sep_token = '[SEP]' pad_token = '[PAD]' cls_token = '[CLS]' mask_token = '[MASK]' tokenize_chinese_chars = True strip_accents = None **kwargs ) → Dict[str, List[List[int]]]

Parameters

  • questions (str or List[str]) — The questions to be encoded. You can specify one question for many passages. In this case, the question will be duplicated like [questions] * n_passages. Otherwise you have to specify as many questions as in titles or texts.
  • titles (str or List[str]) — The passages titles to be encoded. This can be a string or a list of strings if there are several passages.
  • texts (str or List[str]) — The passages texts to be encoded. This can be a string or a list of strings if there are several passages.
  • padding (bool, str or PaddingStrategy, optional, defaults to False) — Activates and controls padding. Accepts the following values:

    • True or 'longest': Pad to the longest sequence in the batch (or no padding if only a single sequence if provided).
    • 'max_length': Pad to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided.
    • False or 'do_not_pad' (default): No padding (i.e., can output a batch with sequences of different lengths).
  • truncation (bool, str or TruncationStrategy, optional, defaults to False) — Activates and controls truncation. Accepts the following values:

    • True or 'longest_first': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will truncate token by token, removing a token from the longest sequence in the pair if a pair of sequences (or a batch of pairs) is provided.
    • 'only_first': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the first sequence of a pair if a pair of sequences (or a batch of pairs) is provided.
    • 'only_second': Truncate to a maximum length specified with the argument max_length or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the second sequence of a pair if a pair of sequences (or a batch of pairs) is provided.
    • False or 'do_not_truncate' (default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size).
  • max_length (int, optional) — Controls the maximum length to use by one of the truncation/padding parameters.

    If left unset or set to None, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated.

  • return_tensors (str or TensorType, optional) — If set, will return tensors instead of list of python integers. Acceptable values are:

    • 'tf': Return TensorFlow tf.constant objects.
    • 'pt': Return PyTorch torch.Tensor objects.
    • 'np': Return Numpy np.ndarray objects.
  • return_attention_mask (bool, optional) — Whether or not to return the attention mask. If not set, will return the attention mask according to the specific tokenizer’s default, defined by the return_outputs attribute.

    What are attention masks?

Returns

Dict[str, List[List[int]]]

A dictionary with the following keys:

  • input_ids: List of token ids to be fed to a model.
  • attention_mask: List of indices specifying which tokens should be attended to by the model.

Constructs a “fast” DPRReader tokenizer (backed by HuggingFace’s tokenizers library).

DPRReaderTokenizerFast is almost identical to BertTokenizerFast and runs end-to-end tokenization: punctuation splitting and wordpiece. The difference is that is has three inputs strings: question, titles and texts that are combined to be fed to the DPRReader model.

Refer to superclass BertTokenizerFast for usage examples and documentation concerning parameters.

Return a dictionary with the token ids of the input strings and other information to give to .decode_best_spans. It converts the strings of a question and different passages (title and text) in a sequence of IDs (integers), using the tokenizer and vocabulary. The resulting input_ids is a matrix of size (n_passages, sequence_length) with the format:

[CLS] <question token ids> [SEP] <titles ids> [SEP] <texts ids>

DPR specific outputs

class transformers.models.dpr.modeling_dpr.DPRContextEncoderOutput

< >

( pooler_output: FloatTensor hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

  • pooler_output (torch.FloatTensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the context representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed contexts for nearest neighbors queries with questions embeddings.
  • 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 for outputs of DPRQuestionEncoder.

class transformers.models.dpr.modeling_dpr.DPRQuestionEncoderOutput

< >

( pooler_output: FloatTensor hidden_states: typing.Optional[typing.Tuple[torch.FloatTensor]] = None attentions: typing.Optional[typing.Tuple[torch.FloatTensor]] = None )

Parameters

  • pooler_output (torch.FloatTensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the question representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed questions for nearest neighbors queries with context embeddings.
  • 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 for outputs of DPRQuestionEncoder.

class transformers.DPRReaderOutput

< >

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

Parameters

  • start_logits (torch.FloatTensor of shape (n_passages, sequence_length)) — Logits of the start index of the span for each passage.
  • end_logits (torch.FloatTensor of shape (n_passages, sequence_length)) — Logits of the end index of the span for each passage.
  • relevance_logits (torch.FloatTensor of shape (n_passages, )) — Outputs of the QA classifier of the DPRReader that corresponds to the scores of each passage to answer the question, compared to all the other passages.
  • 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 for outputs of DPRQuestionEncoder.

DPRContextEncoder

class transformers.DPRContextEncoder

< >

( config: DPRConfig )

Parameters

  • config (DPRConfig) — 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 DPRContextEncoder transformer outputting pooler outputs as context representations.

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 token_type_ids: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None output_attentions = None output_hidden_states = None return_dict = None ) → transformers.models.dpr.modeling_dpr.DPRContextEncoderOutput or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. To match pretraining, DPR input sequence should be formatted with [CLS] and [SEP] tokens as follows:

    (a) For sequence pairs (for a pair title+text for example):

Returns

transformers.models.dpr.modeling_dpr.DPRContextEncoderOutput or tuple(torch.FloatTensor)

A transformers.models.dpr.modeling_dpr.DPRContextEncoderOutput 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 (DPRConfig) and inputs.

  • pooler_output (torch.FloatTensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the context representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed contexts for nearest neighbors queries with questions embeddings.

  • 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 DPRContextEncoder 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 DPRContextEncoder, DPRContextEncoderTokenizer

>>> tokenizer = DPRContextEncoderTokenizer.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
>>> model = DPRContextEncoder.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
>>> input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
>>> embeddings = model(input_ids).pooler_output

DPRQuestionEncoder

class transformers.DPRQuestionEncoder

< >

( config: DPRConfig )

Parameters

  • config (DPRConfig) — 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 DPRQuestionEncoder transformer outputting pooler outputs as question representations.

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 token_type_ids: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.Tensor] = None output_attentions = None output_hidden_states = None return_dict = None ) → transformers.models.dpr.modeling_dpr.DPRQuestionEncoderOutput or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. To match pretraining, DPR input sequence should be formatted with [CLS] and [SEP] tokens as follows:

    (a) For sequence pairs (for a pair title+text for example):

Returns

transformers.models.dpr.modeling_dpr.DPRQuestionEncoderOutput or tuple(torch.FloatTensor)

A transformers.models.dpr.modeling_dpr.DPRQuestionEncoderOutput 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 (DPRConfig) and inputs.

  • pooler_output (torch.FloatTensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the question representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed questions for nearest neighbors queries with context embeddings.

  • 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 DPRQuestionEncoder 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 DPRQuestionEncoder, DPRQuestionEncoderTokenizer

>>> tokenizer = DPRQuestionEncoderTokenizer.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
>>> model = DPRQuestionEncoder.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
>>> input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
>>> embeddings = model(input_ids).pooler_output

DPRReader

class transformers.DPRReader

< >

( config: DPRConfig )

Parameters

  • config (DPRConfig) — 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 DPRReader transformer outputting span predictions.

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 inputs_embeds: typing.Optional[torch.Tensor] = None output_attentions: bool = None output_hidden_states: bool = None return_dict = None ) → transformers.models.dpr.modeling_dpr.DPRReaderOutput or tuple(torch.FloatTensor)

Parameters

  • input_ids (Tuple[torch.LongTensor] of shapes (n_passages, sequence_length)) — Indices of input sequence tokens in the vocabulary. It has to be a sequence triplet with 1) the question and 2) the passages titles and 3) the passages texts To match pretraining, DPR input_ids sequence should be formatted with [CLS] and [SEP] with the format:

    [CLS] <question token ids> [SEP] <titles ids> [SEP] <texts ids>

    DPR is a model with absolute position embeddings so it’s usually advised to pad the inputs on the right rather than the left.

    Indices can be obtained using DPRReaderTokenizer. See this class documentation for more details.

    What are input IDs?

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

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

    What are attention masks?

  • inputs_embeds (torch.FloatTensor of shape (n_passages, 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.dpr.modeling_dpr.DPRReaderOutput or tuple(torch.FloatTensor)

A transformers.models.dpr.modeling_dpr.DPRReaderOutput 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 (DPRConfig) and inputs.

  • start_logits (torch.FloatTensor of shape (n_passages, sequence_length)) — Logits of the start index of the span for each passage.

  • end_logits (torch.FloatTensor of shape (n_passages, sequence_length)) — Logits of the end index of the span for each passage.

  • relevance_logits (torch.FloatTensor of shape (n_passages, )) — Outputs of the QA classifier of the DPRReader that corresponds to the scores of each passage to answer the question, compared to all the other passages.

  • 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 DPRReader 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 DPRReader, DPRReaderTokenizer

>>> tokenizer = DPRReaderTokenizer.from_pretrained("facebook/dpr-reader-single-nq-base")
>>> model = DPRReader.from_pretrained("facebook/dpr-reader-single-nq-base")
>>> encoded_inputs = tokenizer(
...     questions=["What is love ?"],
...     titles=["Haddaway"],
...     texts=["'What Is Love' is a song recorded by the artist Haddaway"],
...     return_tensors="pt",
... )
>>> outputs = model(**encoded_inputs)
>>> start_logits = outputs.start_logits
>>> end_logits = outputs.end_logits
>>> relevance_logits = outputs.relevance_logits

TFDPRContextEncoder

class transformers.TFDPRContextEncoder

< >

( *args **kwargs )

Parameters

  • config (DPRConfig) — 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 DPRContextEncoder transformer outputting pooler outputs as context representations.

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

TensorFlow models and layers in transformers accept two formats as input:

  • having all inputs as keyword arguments (like PyTorch models), or
  • having all inputs as a list, tuple or dict in the first positional argument.

The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like model.fit() things should “just work” for you - just pass your inputs and labels in any format that model.fit() supports! If, however, you want to use the second format outside of Keras methods like fit() and predict(), such as when creating your own layers or models with the Keras Functional API, there are three possibilities you can use to gather all the input Tensors in the first positional argument:

  • a single Tensor with input_ids only and nothing else: model(input_ids)
  • a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: model([input_ids, attention_mask]) or model([input_ids, attention_mask, token_type_ids])
  • a dictionary with one or several input Tensors associated to the input names given in the docstring: model({"input_ids": input_ids, "token_type_ids": token_type_ids})

Note that when creating models and layers with subclassing then you don’t need to worry about any of this, as you can just pass inputs like you would to any other Python function!

call

< >

( input_ids = None attention_mask: typing.Optional[tensorflow.python.framework.ops.Tensor] = None token_type_ids: typing.Optional[tensorflow.python.framework.ops.Tensor] = None inputs_embeds: typing.Optional[tensorflow.python.framework.ops.Tensor] = None output_attentions = None output_hidden_states = None return_dict = None training: bool = False ) → transformers.models.dpr.modeling_tf_dpr.TFDPRContextEncoderOutput or tuple(tf.Tensor)

Parameters

  • input_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. To match pretraining, DPR input sequence should be formatted with [CLS] and [SEP] tokens as follows:

    (a) For sequence pairs (for a pair title+text for example):

Returns

transformers.models.dpr.modeling_tf_dpr.TFDPRContextEncoderOutput or tuple(tf.Tensor)

A transformers.models.dpr.modeling_tf_dpr.TFDPRContextEncoderOutput 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 (DPRConfig) and inputs.

  • pooler_output (tf.Tensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the context representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed contexts for nearest neighbors queries with questions embeddings.

  • 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 TFDPRContextEncoder 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 TFDPRContextEncoder, DPRContextEncoderTokenizer

>>> tokenizer = DPRContextEncoderTokenizer.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
>>> model = TFDPRContextEncoder.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base", from_pt=True)
>>> input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="tf")["input_ids"]
>>> embeddings = model(input_ids).pooler_output

TFDPRQuestionEncoder

class transformers.TFDPRQuestionEncoder

< >

( *args **kwargs )

Parameters

  • config (DPRConfig) — 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 DPRQuestionEncoder transformer outputting pooler outputs as question representations.

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

TensorFlow models and layers in transformers accept two formats as input:

  • having all inputs as keyword arguments (like PyTorch models), or
  • having all inputs as a list, tuple or dict in the first positional argument.

The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like model.fit() things should “just work” for you - just pass your inputs and labels in any format that model.fit() supports! If, however, you want to use the second format outside of Keras methods like fit() and predict(), such as when creating your own layers or models with the Keras Functional API, there are three possibilities you can use to gather all the input Tensors in the first positional argument:

  • a single Tensor with input_ids only and nothing else: model(input_ids)
  • a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: model([input_ids, attention_mask]) or model([input_ids, attention_mask, token_type_ids])
  • a dictionary with one or several input Tensors associated to the input names given in the docstring: model({"input_ids": input_ids, "token_type_ids": token_type_ids})

Note that when creating models and layers with subclassing then you don’t need to worry about any of this, as you can just pass inputs like you would to any other Python function!

call

< >

( input_ids = None attention_mask: typing.Optional[tensorflow.python.framework.ops.Tensor] = None token_type_ids: typing.Optional[tensorflow.python.framework.ops.Tensor] = None inputs_embeds: typing.Optional[tensorflow.python.framework.ops.Tensor] = None output_attentions = None output_hidden_states = None return_dict = None training: bool = False ) → transformers.models.dpr.modeling_tf_dpr.TFDPRQuestionEncoderOutput or tuple(tf.Tensor)

Parameters

  • input_ids (Numpy array or tf.Tensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. To match pretraining, DPR input sequence should be formatted with [CLS] and [SEP] tokens as follows:

    (a) For sequence pairs (for a pair title+text for example):

Returns

transformers.models.dpr.modeling_tf_dpr.TFDPRQuestionEncoderOutput or tuple(tf.Tensor)

A transformers.models.dpr.modeling_tf_dpr.TFDPRQuestionEncoderOutput 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 (DPRConfig) and inputs.

  • pooler_output (tf.Tensor of shape (batch_size, embeddings_size)) — The DPR encoder outputs the pooler_output that corresponds to the question representation. Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer. This output is to be used to embed questions for nearest neighbors queries with context embeddings.

  • 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 TFDPRQuestionEncoder 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 TFDPRQuestionEncoder, DPRQuestionEncoderTokenizer

>>> tokenizer = DPRQuestionEncoderTokenizer.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
>>> model = TFDPRQuestionEncoder.from_pretrained("facebook/dpr-question_encoder-single-nq-base", from_pt=True)
>>> input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="tf")["input_ids"]
>>> embeddings = model(input_ids).pooler_output

TFDPRReader

class transformers.TFDPRReader

< >

( *args **kwargs )

Parameters

  • config (DPRConfig) — 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 DPRReader transformer outputting span predictions.

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

TensorFlow models and layers in transformers accept two formats as input:

  • having all inputs as keyword arguments (like PyTorch models), or
  • having all inputs as a list, tuple or dict in the first positional argument.

The reason the second format is supported is that Keras methods prefer this format when passing inputs to models and layers. Because of this support, when using methods like model.fit() things should “just work” for you - just pass your inputs and labels in any format that model.fit() supports! If, however, you want to use the second format outside of Keras methods like fit() and predict(), such as when creating your own layers or models with the Keras Functional API, there are three possibilities you can use to gather all the input Tensors in the first positional argument:

  • a single Tensor with input_ids only and nothing else: model(input_ids)
  • a list of varying length with one or several input Tensors IN THE ORDER given in the docstring: model([input_ids, attention_mask]) or model([input_ids, attention_mask, token_type_ids])
  • a dictionary with one or several input Tensors associated to the input names given in the docstring: model({"input_ids": input_ids, "token_type_ids": token_type_ids})

Note that when creating models and layers with subclassing then you don’t need to worry about any of this, as you can just pass inputs like you would to any other Python function!

call

< >

( input_ids = None attention_mask: typing.Optional[tensorflow.python.framework.ops.Tensor] = None inputs_embeds: typing.Optional[tensorflow.python.framework.ops.Tensor] = None output_attentions: bool = None output_hidden_states: bool = None return_dict = None training: bool = False ) → transformers.models.dpr.modeling_tf_dpr.TFDPRReaderOutput or tuple(tf.Tensor)

Parameters

  • input_ids (Numpy array or tf.Tensor of shapes (n_passages, sequence_length)) — Indices of input sequence tokens in the vocabulary. It has to be a sequence triplet with 1) the question and 2) the passages titles and 3) the passages texts To match pretraining, DPR input_ids sequence should be formatted with [CLS] and [SEP] with the format:

    [CLS] <question token ids> [SEP] <titles ids> [SEP] <texts ids>

    DPR is a model with absolute position embeddings so it’s usually advised to pad the inputs on the right rather than the left.

    Indices can be obtained using DPRReaderTokenizer. See this class documentation for more details.

  • attention_mask (Numpy array or tf.Tensor of shape (n_passages, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

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

    What are attention masks?

  • inputs_embeds (Numpy array or tf.Tensor of shape (n_passages, 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_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the config will be used instead.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. This argument can be used in eager mode, in graph mode the value will always be set to True.
  • training (bool, optional, defaults to False) — Whether or not to use the model in training mode (some modules like dropout modules have different behaviors between training and evaluation).

Returns

transformers.models.dpr.modeling_tf_dpr.TFDPRReaderOutput or tuple(tf.Tensor)

A transformers.models.dpr.modeling_tf_dpr.TFDPRReaderOutput 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 (DPRConfig) and inputs.

  • start_logits (tf.Tensor of shape (n_passages, sequence_length)) — Logits of the start index of the span for each passage.

  • end_logits (tf.Tensor of shape (n_passages, sequence_length)) — Logits of the end index of the span for each passage.

  • relevance_logits (tf.Tensor of shape (n_passages, )) — Outputs of the QA classifier of the DPRReader that corresponds to the scores of each passage to answer the question, compared to all the other passages.

  • 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(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 TFDPRReader 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 TFDPRReader, DPRReaderTokenizer

>>> tokenizer = DPRReaderTokenizer.from_pretrained("facebook/dpr-reader-single-nq-base")
>>> model = TFDPRReader.from_pretrained("facebook/dpr-reader-single-nq-base", from_pt=True)
>>> encoded_inputs = tokenizer(
...     questions=["What is love ?"],
...     titles=["Haddaway"],
...     texts=["'What Is Love' is a song recorded by the artist Haddaway"],
...     return_tensors="tf",
... )
>>> outputs = model(encoded_inputs)
>>> start_logits = outputs.start_logits
>>> end_logits = outputs.end_logits
>>> relevance_logits = outputs.relevance_logits
←DistilBERT ELECTRA→