modeling

class GAUAlphaModel(vocab_size=12000, hidden_size=768, intermediate_size=1536, num_hidden_layers=24, max_position_embeddings=512, type_vocab_size=2, attention_key_size=128, norm_eps=1e-12, pad_token_id=0, hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, hidden_act='swish', use_bias=False, normalization='softmax_plus', attention_scale=True)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

The bare GAUAlpha 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.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 GAUAlphaModel. 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 GAUAlphaModel.

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

  • 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.

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

  • 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 2.

  • attention_key_size (int, optional) – The dimensionality of the key used in the gau layer. Defaults to 128.

  • norm_eps (float, optional) – The epsilon value used in the normalization layer. Defaults to 1e-12.

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

  • 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 gau in all encoder layers to drop some attention target. Defaults to 0.1.

  • hidden_act (str, optional) – The activation function used in gau layer. Defaults to swish.

  • use_bias (bool, optional) – Whether or not use bias. Defaults to False.

  • normalization (str, optional) – The normalization method used in gau layer. Defaults to softmax_plus.

  • attention_scale (bool, optional) – Whether or not to scale the attention scores. Defaults to True.

forward(input_ids=None, attention_mask=None, token_type_ids=None)[source]

The GAUAlphaModel 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. Its data type should be int64 and it has a shape of [batch_size, sequence_length].

  • attention_mask (Tensor, optional) – Mask used in gau 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. It is a tensor with shape broadcasted to [batch_size, 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.

Returns

Returns last_hidden_state (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].

Return type

tuple

Example

import paddle
from paddlenlp.transformers import GAUAlphaModel, GAUAlphaTokenizer

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaModel.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')

inputs = tokenizer("欢迎使用百度飞桨!")
inputs = {k:paddle.to_tensor([v], dtype="int64") for (k, v) in inputs.items()}
last_hidden_state = model(**inputs)
class GAUAlphaForMaskedLM(gau_alpha)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

GAUAlpha Model with a masked language modeling head on top.

Parameters

gau_alpha (GAUAlphaModel) – An instance of GAUAlphaModel.

forward(input_ids, token_type_ids=None, attention_mask=None)[source]
Parameters
Returns

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

Return type

Tensor

Example

import paddle
from paddlenlp.transformers import GAUAlphaForMaskedLM, GAUAlphaTokenizer

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaForMaskedLM.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')

inputs = tokenizer("欢迎使用百度飞桨!")
inputs = {k:paddle.to_tensor([v], dtype="int64") for (k, v) in inputs.items()}

logits = model(**inputs)
print(logits.shape)
# [1, 11, 12000]
class GAUAlphaPretrainedModel(*args, **kwargs)[source]

Bases: paddlenlp.transformers.model_utils.PretrainedModel

An abstract class for pretrained GAU-alpha models. It provides GAU-alpha 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.

init_weights(layer)[source]

Initialization hook

base_model_class

alias of paddlenlp.transformers.gau_alpha.modeling.GAUAlphaModel

class GAUAlphaForSequenceClassification(gau_alpha, num_classes=2, dropout=None)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

GAUAlpha Model with a linear layer on top of the output layer, designed for sequence classification/regression tasks like GLUE tasks.

Parameters
  • gau_alpha (GAUAlphaModel) – An instance of paddlenlp.transformers.GAUAlphaModel.

  • num_classes (int, optional) – The number of classes. Default to 2.

  • dropout (float, optional) – The dropout probability for output of GAUAlpha. If None, use the same value as hidden_dropout_prob of paddlenlp.transformers.GAUAlphaModel instance. Defaults to None.

forward(input_ids, token_type_ids=None, attention_mask=None)[source]
Parameters
Returns

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

Return type

Tensor

Example

import paddle
from paddlenlp.transformers import GAUAlphaForSequenceClassification, GAUAlphaTokenizer

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaForSequenceClassification.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')

inputs = tokenizer("欢迎使用百度飞桨!")
inputs = {k:paddle.to_tensor([v], dtype="int64") for (k, v) in inputs.items()}
logits = model(**inputs)
class GAUAlphaForTokenClassification(gau_alpha, num_classes=2, dropout=None)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

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

Parameters
  • gau_alpha (GAUAlphaModel) – An instance of paddlenlp.transformers.GAUAlphaModel.

  • num_classes (int, optional) – The number of classes. Default to 2.

  • dropout (float, optional) – The dropout probability for output of GAUAlpha. If None, use the same value as hidden_dropout_prob of paddlenlp.transformers.GAUAlphaModel instance. Defaults to None.

forward(input_ids, token_type_ids=None, attention_mask=None)[source]
Parameters
Returns

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

Return type

Tensor

Example

import paddle
from paddlenlp.transformers import GAUAlphaForTokenClassification, GAUAlphaTokenizer

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaForTokenClassification.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')

inputs = tokenizer("欢迎使用百度飞桨!")
inputs = {k:paddle.to_tensor([v], dtype="int64") for (k, v) in inputs.items()}
logits = model(**inputs)
class GAUAlphaForQuestionAnswering(gau_alpha, dropout=None)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

GAUAlpha with a linear layer on top of the hidden-states output to compute span_start_logits and span_end_logits, designed for question-answering tasks like SQuAD.

Parameters
  • gau_alpha (GAUAlphaModel) – An instance of GAUAlphaModel.

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

forward(input_ids, token_type_ids=None, attention_mask=None)[source]

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

Parameters
Returns

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].

Return type

tuple

Example

import paddle
from paddlenlp.transformers import GAUAlphaForQuestionAnswering, GAUAlphaTokenizer

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaForQuestionAnswering.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')

inputs = tokenizer("欢迎使用百度飞桨!")
inputs = {k:paddle.to_tensor([v], dtype="int64") for (k, v) in inputs.items()}
outputs = model(**inputs)

start_logits = outputs[0]
end_logits = outputs[1]
class GAUAlphaForMultipleChoice(gau_alpha, num_choices=2, dropout=None)[source]

Bases: paddlenlp.transformers.gau_alpha.modeling.GAUAlphaPretrainedModel

GAUAlpha Model with a linear layer on top of the hidden-states output layer, designed for multiple choice tasks like RocStories/SWAG tasks.

Parameters
  • gau_alpha (GAUAlphaModel) – An instance of GAUAlphaModel.

  • num_choices (int, optional) – The number of choices. Defaults to 2.

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

forward(input_ids, token_type_ids=None, attention_mask=None)[source]

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

Parameters
  • input_ids (Tensor) – See GAUAlphaModel and shape as [batch_size, num_choice, sequence_length].

  • token_type_ids (Tensor, optional) – See GAUAlphaModel and shape as [batch_size, num_choice, sequence_length].

  • attention_mask (list, optional) – See GAUAlphaModel and shape as [batch_size, num_choice, sequence_length].

Returns

Returns tensor reshaped_logits, a tensor of the multiple choice classification logits. Shape as [batch_size, num_choice] and dtype as float32.

Return type

Tensor

Example

import paddle
from paddlenlp.transformers import GAUAlphaForMultipleChoice, GAUAlphaTokenizer
from paddlenlp.data import Pad

tokenizer = GAUAlphaTokenizer.from_pretrained('chinese_GAU-alpha-char_L-24_H-768')
model = GAUAlphaForMultipleChoice.from_pretrained('chinese_GAU-alpha-char_L-24_H-768', num_choices=2)

data = [
    {
        "question": "如何打开ipad屏幕?",
        "answer1": "按音量按钮。",
        "answer2": "按下锁定按钮。",
        "label": 1,
    },
    {
        "question": "如何缩进一些文本?",
        "answer1": "在开始写之前留一些空格。",
        "answer2": "按空格键。",
        "label": 0,
    },
]

text = []
text_pair = []
for d in data:
    text.append(d["question"])
    text_pair.append(d["answer1"])
    text.append(d["question"])
    text_pair.append(d["answer2"])

inputs = tokenizer(text, text_pair)
input_ids = Pad(axis=0, pad_val=tokenizer.pad_token_id)(inputs["input_ids"])
token_type_ids = Pad(axis=0, pad_val=tokenizer.pad_token_type_id)(inputs["token_type_ids"])

reshaped_logits = model(
    input_ids=paddle.to_tensor(input_ids, dtype="int64"),
    token_type_ids=paddle.to_tensor(token_type_ids, dtype="int64"),
)
print(reshaped_logits.shape)
# [2, 2]