06f1fd69a6
3.7 KiB
This model was published in HF papers on 2020-08-06 and contributed to Hugging Face Transformers on 2021-01-27.
ConvBERT
Overview
The ConvBERT model was proposed in ConvBERT: Improving BERT with Span-based Dynamic Convolution by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan.
The abstract from the paper is the following:
Pre-trained language models like BERT and its variants have recently achieved impressive performance in various natural language understanding tasks. However, BERT heavily relies on the global self-attention block and thus suffers large memory footprint and computation cost. Although all its attention heads query on the whole input sequence for generating the attention map from a global perspective, we observe some heads only need to learn local dependencies, which means the existence of computation redundancy. We therefore propose a novel span-based dynamic convolution to replace these self-attention heads to directly model local dependencies. The novel convolution heads, together with the rest self-attention heads, form a new mixed attention block that is more efficient at both global and local context learning. We equip BERT with this mixed attention design and build a ConvBERT model. Experiments have shown that ConvBERT significantly outperforms BERT and its variants in various downstream tasks, with lower training cost and fewer model parameters. Remarkably, ConvBERTbase model achieves 86.4 GLUE score, 0.7 higher than ELECTRAbase, while using less than 1/4 training cost. Code and pre-trained models will be released.
This model was contributed by abhishek. The original implementation can be found here: https://github.com/yitu-opensource/ConvBert
Usage tips
ConvBERT training tips are similar to those of BERT. For usage tips refer to BERT documentation.
Resources
- Text classification task guide
- Token classification task guide
- Question answering task guide
- Masked language modeling task guide
- Multiple choice task guide
ConvBertConfig
autodoc ConvBertConfig
ConvBertTokenizer
autodoc ConvBertTokenizer - get_special_tokens_mask - save_vocabulary
ConvBertTokenizerFast
autodoc ConvBertTokenizerFast
ConvBertModel
autodoc ConvBertModel - forward
ConvBertForMaskedLM
autodoc ConvBertForMaskedLM - forward
ConvBertForSequenceClassification
autodoc ConvBertForSequenceClassification - forward
ConvBertForMultipleChoice
autodoc ConvBertForMultipleChoice - forward
ConvBertForTokenClassification
autodoc ConvBertForTokenClassification - forward
ConvBertForQuestionAnswering
autodoc ConvBertForQuestionAnswering - forward