Pytorch codes for the paper Semantics-aware BERT for Language Understanding in AAAI 2020
*working in progress
(Our experiment environment for reference)
Python 3.6+ PyTorch (1.0.0) AllenNLP (0.8.1)
GLUE data can be downloaded from GLUE data by running this script and unpack it to directory glue_data. We provide an example data sample in glue_data/MNLI to show how SemBERT works.
This repo shows the example implementation of SemBERT for NLU tasks.
We basically used the pre-trained BERT uncased models so do not forget to pass the parameter --do_lower_case
An example script is
CUDA_VISIBLE_DEVICES=0 \
python run_classifier.py \
--data_dir glue_data/MNLI/ \
--eval_batch_size 32 \
--max_seq_length 200 \
--bert_model bert-base-uncased \
--do_lower_case \
--task_name mnli \
--do_train \
--do_eval \
--do_predict \
--output_dir glue/base_mnli \
--learning_rate 3e-5
The output pred file can be directly used for GLUE online submission and evaluation.
We provde two kinds of semantic labeling method,
-
online: each word sequence are passed to label module to obtain the tags which could be used for online prediction. This would be time-consuming for large corpus. See tag_model/tagging.py
If you want to use the online one, please specify the
--tagger_path
parameter in the run.py file. -
offline: the current one that pre-process the datasets and save them for later loading for training and evaluation. See tag_model/tagger_offline.py
Our labeled data can be downloaded here for quick start.
https://drive.google.com/file/d/1B-_IRWRvR67eLdvT6bM0b2OiyvySkO-x/view?usp=sharing
Note this repo is based on the offline version, so that the column id/index in the data-processor would be slightly different from the original, which is like this:
text_a = line[-3] text_b = line[-2] label = line[-1]
If you use the original data instead of our preprocessed one by tag_model/tagger_offline.py, please modify the index according to the dataset structure.
The SRL model in this implementation used the ELMo-based SRL model from AllenNLP.
Recently, there is a new BERT-based model, which is a nice alternative.
Please kindly cite this paper in your publications if it helps your research:
@inproceedings{zhang2020SemBERT,
title={Semantics-aware {BERT} for language understanding},
author={Zhang, Zhuosheng and Wu, Yuwei and Zhao, Hai and Li, Zuchao and Zhang, Shuailiang and Zhou, Xi and Zhou, Xiang},
booktitle={the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-2020)},
year={2020}
}