modeling

class SqueezeBertModel(vocab_size, embedding_size, hidden_size, num_hidden_layers, num_attention_heads, intermediate_size, hidden_act, hidden_dropout_prob, attention_probs_dropout_prob, max_position_embeddings, type_vocab_size, q_groups, k_groups, v_groups, output_groups, intermediate_groups, post_attention_groups, initializer_range=0.02, layer_norm_eps=1e-12, **kwargs)

基类：paddlenlp.transformers.squeezebert.modeling.SqueezeBertPreTrainedModel

参数

• vocab_size (int) -- Vocabulary size of inputs_ids in SqueezeBertModel. Also is the vocab size of token embedding matrix. Defines the number of different tokens that can be represented by the inputs_ids passed when calling BertModel.

• hidden_size (int, optional) -- Dimensionality of the embedding layer, encoder layer and pooler layer. Defaults to 768.

• num_hidden_layers (int, optional) -- Number of hidden layers in the Transformer encoder. Defaults to 12.

• num_attention_heads (int, optional) -- Number of attention heads for each attention layer in the Transformer encoder. Defaults to 12.

• intermediate_size (int, optional) -- Output chans for intermediate layer.

• hidden_act (str, optional) -- The non-linear activation function in the feed-forward layer. "gelu", "relu" and any other paddle supported activation functions are supported. Defaults to "gelu".

• hidden_dropout_prob (float, optional) -- The dropout probability for all fully connected layers in the embeddings and encoder. Defaults to 0.1.

• attention_probs_dropout_prob (float, optional) -- The dropout probability used in MultiHeadAttention in all encoder layers to drop some attention target. Defaults to 0.1.

• max_position_embeddings (int, optional) -- The maximum value of the dimensionality of position encoding, which dictates the maximum supported length of an input sequence. Defaults to 512.

• type_vocab_size (int, optional) -- The vocabulary size of token_type_ids. Defaults to 16.

• q_groups (int) -- number of query groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• k_groups (int) -- number of key groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• v_groups (int) -- number of value groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• output_groups (int) -- number of output groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• intermediate_groups (int) -- number of intermediate groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• post_groups (int) -- number of post groups for all layers in the BertModule. (eventually we could change the interface to allow different groups for different layers)

• initializer_range (float, optional) --

  The standard deviation of the normal initializer. Defaults to 0.02. .. note:

  A normal_initializer initializes weight matrices as normal distributions.
  See :meth:`BertPretrainedModel.init_weights()` for how weights are initialized in `BertModel`.
  

get_input_embeddings()

get input embedding of model

返回

embedding of model

返回类型

nn.Embedding

set_input_embeddings(new_embeddings)

set new input embedding for model

参数

value (Embedding) -- the new embedding of model

引发

NotImplementedError -- Model has not implement set_input_embeddings method

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

The forward method, overrides the __call__() special method. :param input_ids: Indices of input sequence tokens in the vocabulary. They are

numerical representations of tokens that build the input sequence. Its data type should be int64 and it has a shape of [batch_size, sequence_length].

参数

• attention_mask (Tensor, optional) -- Mask used in multi-head attention to avoid performing attention on to some unwanted positions, usually the paddings or the subsequent positions. Its data type can be int, float and bool. If its data type is int, the values should be either 0 or 1. - 1 for tokens that not masked, - 0 for tokens that masked. It is a tensor with shape broadcasted to [batch_size, num_attention_heads, sequence_length, sequence_length]. Defaults to None, which means nothing needed to be prevented attention to.

• token_type_ids (Tensor, optional) -- Segment token indices to indicate different portions of the inputs. Selected in the range [0, type_vocab_size - 1]. If type_vocab_size is 2, which means the inputs have two portions. Indices can either be 0 or 1: - 0 corresponds to a sentence A token, - 1 corresponds to a sentence B token. Its data type should be int64 and it has a shape of [batch_size, sequence_length]. Defaults to None, which means we don't add segment embeddings.

• position_ids (Tensor, optional) -- Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, max_position_embeddings - 1]. Shape as (batch_size, num_tokens) and dtype as int64. Defaults to None.

• output_attentions (bool, optional) -- Whether to return the attention_weight of each hidden layers. Defaults to False.

• output_hidden_states (bool, optional) -- Whether to return the output of each hidden layers. Defaults to False.

返回

Returns tuple (sequence_output, pooled_output) with (encoder_outputs, encoder_attentions) by optional. With the fields: - sequence_output (Tensor):

Sequence of hidden-states at the last layer of the model. It's data type should be float32 and its shape is [batch_size, sequence_length, hidden_size].

• pooled_output (Tensor):

  The output of first token ([CLS]) in sequence. We "pool" the model by simply taking the hidden state corresponding to the first token. Its data type should be float32 and its shape is [batch_size, hidden_size].

• encoder_outputs (List(Tensor)):

  A list of Tensor containing hidden-states of the model at each hidden layer in the Transformer encoder. The length of the list is num_hidden_layers + 1 (Embedding Layer output). Each Tensor has a data type of float32 and its shape is [batch_size, sequence_length, hidden_size].

返回类型

tuple

class SqueezeBertForSequenceClassification(squeezebert, num_classes=2, dropout=None)

基类：paddlenlp.transformers.squeezebert.modeling.SqueezeBertPreTrainedModel

SqueezeBert Model with a sequence classification/regression head on top (a linear layer on top of the pooled output) e.g. for GLUE tasks. :param squeezebert: An instance of SqueezeBert. :type squeezebert: SqueezeBertModel :param num_classes: The number of classes. Defaults to 2. :type num_classes: int, optional :param dropout: The dropout probability for output of SqueezeBertModel.

If None, use the same value as hidden_dropout_prob of SqueezeBertModel instance squeezebert. Defaults to None.

forward(input_ids, token_type_ids=None, position_ids=None, attention_mask=None)

The SqueezeBertForSequenceClassification forward method, overrides the __call__() special method. :param input_ids: See SqueezeBertModel. :type input_ids: Tensor :param token_type_ids: See SqueezeBertModel. :type token_type_ids: Tensor, optional :param position_ids: See SqueezeBertModel. :type position_ids: Tensor, optional :param attention_mask: See SqueezeBertModel. :type attention_mask: list, optional

返回

Returns tensor logits, a tensor of the input text classification logits. Shape as [batch_size, num_classes] and dtype as float32.

返回类型

Tensor

class SqueezeBertForTokenClassification(squeezebert, num_classes=2, dropout=None)

基类：paddlenlp.transformers.squeezebert.modeling.SqueezeBertPreTrainedModel

SqueezeBert 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. :param squeezebert: An instance of SqueezeBertModel. :type squeezebert: SqueezeBertModel :param num_classes: The number of classes. Defaults to 2. :type num_classes: int, optional :param dropout: The dropout probability for output of squeezebert.

If None, use the same value as hidden_dropout_prob of SqueezeBert instance squeezebert. Defaults to None.

forward(input_ids, token_type_ids=None, position_ids=None, attention_mask=None)

The SqueezeBertForTokenClassification forward method, overrides the __call__() special method. :param input_ids: See SqueezeBertModel. :type input_ids: Tensor :param token_type_ids: See SqueezeBertModel. :type token_type_ids: Tensor, optional :param position_ids: See SqueezeBertModel. :type position_ids: Tensor, optional :param attention_mask: See SqueezeBertModel. :type attention_mask: list, optional

返回

Returns tensor logits, a tensor of the input token classification logits. Shape as [batch_size, sequence_length, num_classes] and dtype as float32.

返回类型

Tensor

class SqueezeBertForQuestionAnswering(squeezebert, dropout=None)

基类：paddlenlp.transformers.squeezebert.modeling.SqueezeBertPreTrainedModel

SqueezeBert 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). :param squeezebert: An instance of SqueezeBertModel. :type squeezebert: SqueezeBertModel :param dropout: The dropout probability for output of SqueezeBert.

If None, use the same value as hidden_dropout_prob of SqueezeBertModel instance squeezebert. Defaults to None.

forward(input_ids, token_type_ids=None)

The SqueezeBertForQuestionAnswering forward method, overrides the __call__() special method. :param input_ids: See SqueezeBertModel. :type input_ids: Tensor :param token_type_ids: See SqueezeBertModel. :type token_type_ids: Tensor, optional

返回

Returns tuple (start_logits, end_logits). With the fields: - start_logits (Tensor):

A tensor of the input token classification logits, indicates the start position of the labelled span. Its data type should be float32 and its shape is [batch_size, sequence_length].

• end_logits (Tensor):

  A tensor of the input token classification logits, indicates the end position of the labelled span. Its data type should be float32 and its shape is [batch_size, sequence_length].

返回类型

tuple