BERT 调用函数中的关键字参数

Keyword arguments in BERT call function

在 HuggingFace TensorFlow 2.0 BERT 库中,documentation 指出:

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.

我正在尝试使用这两个中的第一个来调用我创建的 BERT 模型:

from transformers import BertTokenizer, TFBertModel
import tensorflow as tf

bert_tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
bert_model = TFBertModel.from_pretrained('bert-base-uncased')

text = ['This is a sentence', 
        'The sky is blue and the grass is green', 
        'More words are here']
labels = [0, 1, 0]
tokenized_text = bert_tokenizer.batch_encode_plus(batch_text_or_text_pairs=text,
                                                  pad_to_max_length=True,
                                                  return_tensors='tf')
dataset = tf.data.Dataset.from_tensor_slices((tokenized_text['input_ids'],
                                              tokenized_text['attention_mask'],
                                              tokenized_text['token_type_ids'],
                                              tf.constant(labels))).batch(3)
sample = next(iter(dataset))

result1 = bert_model(inputs=(sample[0], sample[1], sample[2]))  # works fine
result2 = bert_model(inputs={'input_ids': sample[0], 
                             'attention_mask': sample[1], 
                             'token_type_ids': sample[2]})  # also fine
result3 = bert_model(input_ids=sample[0], 
                     attention_mask=sample[1], 
                     token_type_ids=sample[2])  # raises an error

但是当我执行最后一行时,出现错误:

TypeError: __call__() missing 1 required positional argument: 'inputs'

有人可以解释一下如何正确使用输入的关键字参数样式吗?

似乎在内部,他们将 inputs 解释为 input_ids,如果您不将多个张量作为第一个参数。您可以在 TFBertModel 中看到它,然后寻找 TFBertMainLayercall 函数。

对我来说,如果我执行以下操作,我会得到与 result1result2 完全相同的结果:

result3 = bert_model(inputs=sample[0], 
                     attention_mask=sample[1], 
                     token_type_ids=sample[2])

或者,您也可以只删除 inputs=,同样有效。