Hugging Face's logo

yangheng
/
deberta-v3-large-absa

Transformers
PyTorch
English
deberta-v2
aspect-based-sentiment-analysis
lcf-bert
Inference Endpoints
Model card Files Community
yangheng commited on
Commit
8d792a4
•
1 Parent(s): 7c09b54

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +105 -0
README.md CHANGED
@@ -1,3 +1,108 @@
1
  ---
 
 
 
 
 
2
  license: mit
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - aspect-based-sentiment-analysis
6
+ - lcf-bert
7
  license: mit
8
+ datasets:
9
+ - laptop14 (w/ augmentation)
10
+ - restaurant14 (w/ augmentation)
11
+ - restaurant16 (w/ augmentation)
12
+ - ACL-Twitter (w/ augmentation)
13
+ - MAMS (w/ augmentation)
14
+ - Television (w/ augmentation)
15
+ - TShirt (w/ augmentation)
16
+ - Yelp (w/ augmentation)
17
+ metrics:
18
+ - accuracy
19
+ - macro-f1
20
+
21
  ---
22
+
23
+ # DeBERTa for aspect-based sentiment analysis
24
+ The `deberta-v3-large-absa` model for aspect-based sentiment analysis, trained with English datasets from [ABSADatasets](https://github.com/yangheng95/ABSADatasets).
25
+
26
+ ## Training Model
27
+ This model is trained based on the FAST-LSA-T model with `microsoft/deberta-v3-large`, which comes from [PyABSA](https://github.com/yangheng95/PyABSA).
28
+ To track state-of-the-art models, please see [PyASBA](https://github.com/yangheng95/PyABSA).
29
+
30
+ ## Usage
31
+ ```python3
32
+ from transformers import AutoTokenizer, AutoModel
33
+
34
+ tokenizer = AutoTokenizer.from_pretrained("yangheng/deberta-v3-large-absa")
35
+ model = AutoModel.from_pretrained("yangheng/deberta-v3-large-absa")
36
+
37
+ inputs = tokenizer("good product especially video and audio quality fantastic.", return_tensors="pt")
38
+ outputs = model(**inputs)
39
+ ```
40
+
41
+ ## Example in PyASBA
42
+ An [example](https://github.com/yangheng95/PyABSA/blob/release/demos/aspect_polarity_classification/train_apc_multilingual.py) for using FAST-LSA-T in PyASBA
43
+
44
+ ## Datasets
45
+ This model is fine-tuned with 180k examples for the ABSA dataset (including augmented data). Training dataset files:
46
+ ```
47
+ loading: integrated_datasets/apc_datasets/SemEval/laptop14/Laptops_Train.xml.seg
48
+ loading: integrated_datasets/apc_datasets/SemEval/laptop14/0.cross_boost.fast_lcf_bert_Laptop14_deberta-v3-base.train.augment
49
+ loading: integrated_datasets/apc_datasets/SemEval/laptop14/1.cross_boost.fast_lcf_bert_Laptop14_deberta-v3-base.train.augment
50
+ loading: integrated_datasets/apc_datasets/SemEval/laptop14/2.cross_boost.fast_lcf_bert_Laptop14_deberta-v3-base.train.augment
51
+ loading: integrated_datasets/apc_datasets/SemEval/laptop14/3.cross_boost.fast_lcf_bert_Laptop14_deberta-v3-base.train.augment
52
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant14/Restaurants_Train.xml.seg
53
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant14/0.cross_boost.fast_lcf_bert_Restaurant14_deberta-v3-base.train.augment
54
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant14/1.cross_boost.fast_lcf_bert_Restaurant14_deberta-v3-base.train.augment
55
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant14/2.cross_boost.fast_lcf_bert_Restaurant14_deberta-v3-base.train.augment
56
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant14/3.cross_boost.fast_lcf_bert_Restaurant14_deberta-v3-base.train.augment
57
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant16/restaurant_train.raw
58
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant16/0.cross_boost.fast_lcf_bert_Restaurant16_deberta-v3-base.train.augment
59
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant16/1.cross_boost.fast_lcf_bert_Restaurant16_deberta-v3-base.train.augment
60
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant16/2.cross_boost.fast_lcf_bert_Restaurant16_deberta-v3-base.train.augment
61
+ loading: integrated_datasets/apc_datasets/SemEval/restaurant16/3.cross_boost.fast_lcf_bert_Restaurant16_deberta-v3-base.train.augment
62
+ loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/train.raw
63
+ loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/0.cross_boost.fast_lcf_bert_Twitter_deberta-v3-base.train.augment
64
+ loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/1.cross_boost.fast_lcf_bert_Twitter_deberta-v3-base.train.augment
65
+ loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/2.cross_boost.fast_lcf_bert_Twitter_deberta-v3-base.train.augment
66
+ loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/3.cross_boost.fast_lcf_bert_Twitter_deberta-v3-base.train.augment
67
+ loading: integrated_datasets/apc_datasets/MAMS/train.xml.dat
68
+ loading: integrated_datasets/apc_datasets/MAMS/0.cross_boost.fast_lcf_bert_MAMS_deberta-v3-base.train.augment
69
+ loading: integrated_datasets/apc_datasets/MAMS/1.cross_boost.fast_lcf_bert_MAMS_deberta-v3-base.train.augment
70
+ loading: integrated_datasets/apc_datasets/MAMS/2.cross_boost.fast_lcf_bert_MAMS_deberta-v3-base.train.augment
71
+ loading: integrated_datasets/apc_datasets/MAMS/3.cross_boost.fast_lcf_bert_MAMS_deberta-v3-base.train.augment
72
+ loading: integrated_datasets/apc_datasets/Television/Television_Train.xml.seg
73
+ loading: integrated_datasets/apc_datasets/Television/0.cross_boost.fast_lcf_bert_Television_deberta-v3-base.train.augment
74
+ loading: integrated_datasets/apc_datasets/Television/1.cross_boost.fast_lcf_bert_Television_deberta-v3-base.train.augment
75
+ loading: integrated_datasets/apc_datasets/Television/2.cross_boost.fast_lcf_bert_Television_deberta-v3-base.train.augment
76
+ loading: integrated_datasets/apc_datasets/Television/3.cross_boost.fast_lcf_bert_Television_deberta-v3-base.train.augment
77
+ loading: integrated_datasets/apc_datasets/TShirt/Menstshirt_Train.xml.seg
78
+ loading: integrated_datasets/apc_datasets/TShirt/0.cross_boost.fast_lcf_bert_TShirt_deberta-v3-base.train.augment
79
+ loading: integrated_datasets/apc_datasets/TShirt/1.cross_boost.fast_lcf_bert_TShirt_deberta-v3-base.train.augment
80
+ loading: integrated_datasets/apc_datasets/TShirt/2.cross_boost.fast_lcf_bert_TShirt_deberta-v3-base.train.augment
81
+ loading: integrated_datasets/apc_datasets/TShirt/3.cross_boost.fast_lcf_bert_TShirt_deberta-v3-base.train.augment
82
+ loading: integrated_datasets/apc_datasets/Yelp/yelp.train.txt
83
+ loading: integrated_datasets/apc_datasets/Yelp/0.cross_boost.fast_lcf_bert_Yelp_deberta-v3-base.train.augment
84
+ loading: integrated_datasets/apc_datasets/Yelp/1.cross_boost.fast_lcf_bert_Yelp_deberta-v3-base.train.augment
85
+ loading: integrated_datasets/apc_datasets/Yelp/2.cross_boost.fast_lcf_bert_Yelp_deberta-v3-base.train.augment
86
+ loading: integrated_datasets/apc_datasets/Yelp/3.cross_boost.fast_lcf_bert_Yelp_deberta-v3-base.train.augment
87
+
88
+ ```
89
+ If you use this model in your research, please cite our paper:
90
+ ```
91
+ @article{YangZMT21,
92
+ author = {Heng Yang and
93
+ Biqing Zeng and
94
+ Mayi Xu and
95
+ Tianxing Wang},
96
+ title = {Back to Reality: Leveraging Pattern-driven Modeling to Enable Affordable
97
+ Sentiment Dependency Learning},
98
+ journal = {CoRR},
99
+ volume = {abs/2110.08604},
100
+ year = {2021},
101
+ url = {https://arxiv.org/abs/2110.08604},
102
+ eprinttype = {arXiv},
103
+ eprint = {2110.08604},
104
+ timestamp = {Fri, 22 Oct 2021 13:33:09 +0200},
105
+ biburl = {https://dblp.org/rec/journals/corr/abs-2110-08604.bib},
106
+ bibsource = {dblp computer science bibliography, https://dblp.org}
107
+ }
108
+ ```