Views
No views yet
base version of ElasticBERT.1>>> from transformers import BertTokenizer as ElasticBertTokenizer
2>>> from models.configuration_elasticbert import ElasticBertConfig
3>>> from models.modeling_elasticbert import ElasticBertForSequenceClassification
4
5>>> num_output_layers = 1
6>>> config = ElasticBertConfig.from_pretrained('fnlp/elasticbert-base', num_output_layers=num_output_layers )
7>>> tokenizer = ElasticBertTokenizer.from_pretrained('fnlp/elasticbert-base')
8>>> model = ElasticBertForSequenceClassification.from_pretrained('fnlp/elasticbert-base', config=config)
9
10>>> input_ids = tokenizer.encode('The actors are fantastic .', return_tensors='pt')
11>>> outputs = model(input_ids)1@article{liu2021elasticbert,
2 author = {Xiangyang Liu and
3 Tianxiang Sun and
4 Junliang He and
5 Lingling Wu and
6 Xinyu Zhang and
7 Hao Jiang and
8 Zhao Cao and
9 Xuanjing Huang and
10 Xipeng Qiu},
11 title = {Towards Efficient {NLP:} {A} Standard Evaluation and {A} Strong Baseline},
12 journal = {CoRR},
13 volume = {abs/2110.07038},
14 year = {2021},
15 url = {https://arxiv.org/abs/2110.07038},
16 eprinttype = {arXiv},
17 eprint = {2110.07038},
18 timestamp = {Fri, 22 Oct 2021 13:33:09 +0200},
19 biburl = {https://dblp.org/rec/journals/corr/abs-2110-07038.bib},
20 bibsource = {dblp computer science bibliography, https://dblp.org}
21}