modeling#

class ErnieCtmPretrainedModel(*args, **kwargs)[source]#

Bases: PretrainedModel

An abstract class for pretrained ErnieCtm models. It provides ErnieCtm related model_config_file, pretrained_init_configuration, resource_files_names, pretrained_resource_files_map, base_model_prefix for downloading

and loading pretrained models.

See PretrainedModel for more details.

config_class#

alias of ErnieCtmConfig

base_model_class#

alias of ErnieCtmModel

class ErnieCtmModel(config: ErnieCtmConfig)[source]#

Bases: ErnieCtmPretrainedModel

The bare ErnieCtm Model transformer outputting raw hidden-states.

This model inherits from PretrainedModel. Refer to the superclass documentation for the generic methods.

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

Parameters:

  • vocab_size (int) – Vocabulary size of inputs_ids in ErnieCtmModel. 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 ErnieCtmModel.

  • embedding_size (int, optional) – Dimensionality of the embedding layer. Defaults to 128.

  • hidden_size (int, optional) – Dimensionality of the encoder layers and the 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) – Dimensionality of the feed-forward (ff) layer in the encoder. Input tensors to ff layers are firstly projected from hidden_size to intermediate_size, and then projected back to hidden_size. Typically intermediate_size is larger than hidden_size. Defaults to 3072.

  • 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 the token_type_ids. Defaults to 16.

  • initializer_range (float, optional) – The standard deviation of the normal initializer for initializing all weight matrices. Defaults to 0.02.

  • pad_token_id (int, optional) – The index of padding token in the token vocabulary. Defaults to 0.

  • use_content_summary (bool, optional) – Whether or not to add content summary tokens. Defaults to True.

  • content_summary_index (int, optional) – The number of the content summary tokens. Only valid when use_content_summary is True. Defaults to 1.

  • cls_num (int, optional) – The number of the CLS tokens. Only valid when use_content_summary is True. Defaults to 2.

get_input_embeddings()[source]#

get input embedding of model

Returns:

embedding of model

Return type:

nn.Embedding

set_input_embeddings(value)[source]#

set new input embedding for model

Parameters:

value (Embedding) – the new embedding of model

Raises:

NotImplementedError – Model has not implement set_input_embeddings method

forward(input_ids=None, token_type_ids=None, position_ids=None, attention_mask=None, inputs_embeds=None, content_clone=False, output_hidden_states=None, output_attentions=None, return_dict=None)[source]#

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

Parameters:

  • input_ids (Tensor) – Indices of input sequence tokens in the vocabulary. They are numerical representations of tokens that build the input sequence. It’s data type should be int64 and has a shape of [batch_size, sequence_length].

  • 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.

  • 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. When the data type is bool, the masked tokens have False values and the others have True values. When the data type is int, the masked tokens have 0 values and the others have 1 values. When the data type is float, the masked tokens have -INF values and the others have 0 values. It is a tensor with shape broadcasted to [batch_size, num_attention_heads, sequence_length, sequence_length]. For example, its shape can be [batch_size, sequence_length], [batch_size, sequence_length, sequence_length], [batch_size, num_attention_heads, sequence_length, sequence_length]. We use whole-word-mask in ERNIE, so the whole word will have the same value. For example, “使用” as a word, “使” and “用” will have the same value. Defaults to None, which means nothing needed to be prevented attention to.

  • inputs_embeds (Tensor, optional) – Optionally, instead of passing input_ids you can choose to directly pass an embedded representation of shape (batch_size, sequence_length, hidden_size). 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. Default to None.

  • content_clone (bool, optional) – Whether the content_output is clone from sequence_output. If set to True, the content_output is clone from sequence_output, which may cause the classification task impact on the sequence labeling task. Defaults to False.

  • output_hidden_states (bool, optional) – Whether to return the hidden states of all layers. Defaults to None.

  • output_attentions (bool, optional) – Whether to return the attentions tensors of all attention layers. Defaults to None. (currently not supported)

  • return_dict (bool, optional) – Whether to return a ModelOutput object. If False, the output will be a tuple of tensors. Defaults to None.

Returns:

Returns tuple (sequence_output, pooled_output, content_output).

With the fields:

  • sequence_output (Tensor):

    Sequence of output at the last layer of the model. Its data type should be float32 and has a shape of [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].

  • content_output (Tensor):

    The output of content summary token ([CLS1] in sequence). Its data type should be float32 and has a shape of [batch_size, hidden_size].

Return type:

tuple

Example

import paddle
from paddlenlp.transformers import ErnieModel, ErnieTokenizer

tokenizer = ErnieTokenizer.from_pretrained('ernie-1.0')
model = ErnieModel.from_pretrained('ernie-1.0')

inputs = tokenizer("Welcome to use PaddlePaddle and PaddleNLP!")
inputs = {k:paddle.to_tensor([v]) for (k, v) in inputs.items()}
sequence_output, pooled_output, content_output = model(**inputs)
class ErnieCtmWordtagModel(config: ErnieCtmConfig)[source]#

Bases: ErnieCtmPretrainedModel

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

Parameters:

  • ernie_ctm (:clss:`ErnieCtmModel`) – An instance of ErnieCtmModel.

  • num_tag (int) – The number of different tags.

  • crf_lr (float) – The learning rate of the crf. Defaults to 100.

forward(input_ids=None, token_type_ids=None, lengths=None, position_ids=None, attention_mask=None, inputs_embeds=None, tag_labels=None, output_hidden_states=None, output_attentions=None, return_dict=None, **kwargs)[source]#
Parameters:

  • input_ids (Tensor) – See ErnieCtmModel.

  • token_type_ids (Tensor, optional) – See ErnieCtmModel.

  • position_ids (Tensor, optional) – See ErnieCtmModel.

  • attention_mask (Tensor, optional) – See ErnieCtmModel.

  • inputs_embeds (Tensor, optional) – See ErnieCtmModel.

  • lengths (Tensor, optional) – The input length. Its dtype is int64 and has a shape of [batch_size]. Defaults to None.

  • tag_labels (Tensor, optional) – The input predicted tensor. Its dtype is float32 and has a shape of [batch_size, sequence_length, num_tags]. Defaults to None.

  • output_hidden_states (bool, optional) – Whether to return the hidden states of all layers. Defaults to None.

  • output_attentions (bool, optional) – Whether to return the attentions tensors of all attention layers. Defaults to None. (currently not supported)

  • return_dict (bool, optional) – Whether to return a ModelOutput object. If False, the output will be a tuple of tensors. Defaults to None.

Returns:

Returns tuple (seq_logits, cls_logits).

With the fields:

  • seq_logits (Tensor):

    A tensor of next sentence prediction logits. Its data type should be float32 and its shape is [batch_size, sequence_length, num_tag].

Return type:

tuple

Example

import paddle
from paddlenlp.transformers import ErnieCtmWordtagModel, ErnieCtmTokenizer

tokenizer = ErnieCtmTokenizer.from_pretrained('ernie-ctm')
model = ErnieCtmWordtagModel.from_pretrained('ernie-ctm', num_tag=2)

inputs = tokenizer("Welcome to use PaddlePaddle and PaddleNLP!")
inputs = {k:paddle.to_tensor([v]) for (k, v) in inputs.items()}
logits = model(**inputs)
class ErnieCtmNptagModel(config: ErnieCtmConfig)[source]#

Bases: ErnieCtmPretrainedModel

ErnieCtmNptag Model with a masked language modeling head on top.

Parameters:

ernie_ctm (:clss:`ErnieCtmModel`) – An instance of ErnieCtmModel.

forward(input_ids=None, token_type_ids=None, attention_mask=None, position_ids=None, inputs_embeds=None, labels=None, output_hidden_states: bool | None = None, output_attentions: bool | None = None, return_dict: bool | None = None)[source]#
Parameters:
Returns:

Returns tensor logits, the scores of masked token prediction. Its data type should be float32 and shape is [batch_size, sequence_length, vocab_size].

Return type:

tuple

Example

import paddle
from paddlenlp.transformers import ErnieCtmNptagModel, ErnieCtmTokenizer

tokenizer = ErnieCtmTokenizer.from_pretrained('ernie-ctm')
model = ErnieCtmNptagModel.from_pretrained('ernie-ctm')

inputs = tokenizer("Welcome to use PaddlePaddle and PaddleNLP!")
inputs = {k:paddle.to_tensor([v]) for (k, v) in inputs.items()}

logits = model(**inputs)
print(logits.shape)
# [1, 45, 23000]
class ErnieCtmForTokenClassification(config: ErnieCtmConfig)[source]#

Bases: ErnieCtmPretrainedModel

ERNIECtm Model with a linear layer on top of the hidden-states output layer, designed for token classification tasks like NER tasks.

Parameters:

  • ernie (ErnieModel) – An instance of ErnieModel.

  • num_tag (int, optional) – The number of classes. Defaults to 2.

  • dropout (float, optional) – The dropout probability for output of ERNIE. If None, use the same value as hidden_dropout_prob of ErnieCtmModel instance ernie. Defaults to None.

forward(input_ids: Tensor, token_type_ids: Tensor | None = None, position_ids: Tensor | None = None, attention_mask: Tensor | None = None, inputs_embeds: Tensor | None = None, labels: Tensor | None = None, output_hidden_states: bool | None = None, output_attentions: bool | None = None, return_dict: bool | None = None)[source]#
Parameters:

  • input_ids (Tensor) – See ErnieCtmModel.

  • token_type_ids (Tensor, optional) – See ErnieCtmModel.

  • position_ids (Tensor, optional) – See ErnieCtmModel.

  • attention_mask (Tensor, optional) – See ErnieCtmModel.

  • inputs_embeds (Tensor, optional) – See ErnieCtmModel.

  • labels (Tensor, optional) – labels for model to compute the loss

Returns:

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

Return type:

Tensor

Example

import paddle
from paddlenlp.transformers import ErnieCtmForTokenClassification, ErnieCtmTokenizer

tokenizer = ErnieCtmTokenizer.from_pretrained('ernie-ctm')
model = ErnieCtmForTokenClassification.from_pretrained('ernie-ctm')

inputs = tokenizer("Welcome to use PaddlePaddle and PaddleNLP!")
inputs = {k:paddle.to_tensor([v]) for (k, v) in inputs.items()}
logits = model(**inputs)