|
--- |
|
language: |
|
- en |
|
tags: |
|
- aspect-based-sentiment-analysis |
|
- PyABSA |
|
license: mit |
|
datasets: |
|
- laptop14 |
|
- restaurant14 |
|
- restaurant16 |
|
- ACL-Twitter |
|
- MAMS |
|
- Television |
|
- TShirt |
|
- Yelp |
|
metrics: |
|
- accuracy |
|
- macro-f1 |
|
widget: |
|
- text: "[CLS] when tables opened up, the manager sat another party before us. [SEP] manager [SEP] " |
|
--- |
|
|
|
|
|
# Note |
|
Please use (yangheng/deberta-v3-base-absa-v1.1)[https://huggingface.co/yangheng/deberta-v3-base-absa-v1.1], which is smaller and has better performance. |
|
This model is training with 30k+ ABSA samples, see [ABSADatasets](https://github.com/yangheng95/ABSADatasets). Yet the test sets are not included in pre-training, so you can use this model for training and benchmarking on common ABSA datasets, e.g., Laptop14, Rest14 datasets. (Except for the Rest15 dataset!) |
|
|
|
# DeBERTa for aspect-based sentiment analysis |
|
The `deberta-v3-large-absa` model for aspect-based sentiment analysis, trained with English datasets from [ABSADatasets](https://github.com/yangheng95/ABSADatasets). |
|
|
|
## Training Model |
|
This model is trained based on the FAST-LCF-BERT model with `microsoft/deberta-v3-large`, which comes from [PyABSA](https://github.com/yangheng95/PyABSA). |
|
To track state-of-the-art models, please see [PyASBA](https://github.com/yangheng95/PyABSA). |
|
|
|
## Usage |
|
```python3 |
|
from transformers import AutoTokenizer, AutoModelForSequenceClassification |
|
|
|
tokenizer = AutoTokenizer.from_pretrained("yangheng/deberta-v3-large-absa-v1.1") |
|
|
|
model = AutoModelForSequenceClassification.from_pretrained("yangheng/deberta-v3-large-absa-v1.1") |
|
``` |
|
|
|
## Example in PyASBA |
|
An [example](https://github.com/yangheng95/PyABSA/blob/release/demos/aspect_polarity_classification/train_apc_multilingual.py) for using FAST-LCF-BERT in PyASBA datasets. |
|
|
|
## Datasets |
|
This model is fine-tuned with 180k examples for the ABSA dataset (including augmented data). Training dataset files: |
|
``` |
|
loading: integrated_datasets/apc_datasets/SemEval/laptop14/Laptops_Train.xml.seg |
|
loading: integrated_datasets/apc_datasets/SemEval/restaurant14/Restaurants_Train.xml.seg |
|
loading: integrated_datasets/apc_datasets/SemEval/restaurant16/restaurant_train.raw |
|
loading: integrated_datasets/apc_datasets/ACL_Twitter/acl-14-short-data/train.raw |
|
loading: integrated_datasets/apc_datasets/MAMS/train.xml.dat |
|
loading: integrated_datasets/apc_datasets/Television/Television_Train.xml.seg |
|
loading: integrated_datasets/apc_datasets/TShirt/Menstshirt_Train.xml.seg |
|
loading: integrated_datasets/apc_datasets/Yelp/yelp.train.txt |
|
|
|
``` |
|
If you use this model in your research, please cite our paper: |
|
``` |
|
@article{YangZMT21, |
|
author = {Heng Yang and |
|
Biqing Zeng and |
|
Mayi Xu and |
|
Tianxing Wang}, |
|
title = {Back to Reality: Leveraging Pattern-driven Modeling to Enable Affordable |
|
Sentiment Dependency Learning}, |
|
journal = {CoRR}, |
|
volume = {abs/2110.08604}, |
|
year = {2021}, |
|
url = {https://arxiv.org/abs/2110.08604}, |
|
eprinttype = {arXiv}, |
|
eprint = {2110.08604}, |
|
timestamp = {Fri, 22 Oct 2021 13:33:09 +0200}, |
|
biburl = {https://dblp.org/rec/journals/corr/abs-2110-08604.bib}, |
|
bibsource = {dblp computer science bibliography, https://dblp.org} |
|
} |
|
``` |
