Trendos/deberta-v3-large-absa-v1.1-onnx

deberta-v3-large-absa-v1.1 (ONNX)

ONNX export of yangheng/deberta-v3-large-absa-v1.1 — a state-of-the-art aspect-based sentiment analysis model built on microsoft/deberta-v3-large and fine-tuned with the PyABSA framework.

• Language: Multilingual (EN, ZH, AR, NL, FR, RU, ES, TR, and more)
• Task: Text Classification / Aspect-Based Sentiment Analysis (ABSA)
• License: MIT
• Original model: yangheng/deberta-v3-large-absa-v1.1

---

What's different here

Same weights, exported to ONNX for faster CPU inference and easier deployment without a PyTorch dependency.

---

Inference

from optimum.onnxruntime import ORTModelForSequenceClassification
from transformers import AutoTokenizer, pipeline

onnxModelDirectory = "./onnx"

tokenizer = AutoTokenizer.from_pretrained(onnxModelDirectory)
onnxModel = ORTModelForSequenceClassification.from_pretrained(onnxModelDirectory)

absa_classifier = pipeline("text-classification", model=onnxModel, tokenizer=tokenizer)

sentence = "The food was exceptional, although the service was a bit slow."

foodSentiment = absa_classifier(sentence, text_pair="food")
serviceSentiment = absa_classifier(sentence, text_pair="service")

print(foodSentiment)   # Positive
print(serviceSentiment) # Negative

---

Output labels

Label	Description
Positive	Positive sentiment toward the aspect
Neutral	Neutral sentiment toward the aspect
Negative	Negative sentiment toward the aspect

---

License

MIT — see LICENSE.

Original work by Heng Yang. Copyright notice must be retained in all copies or substantial portions of the model/software.

---

Citation

@inproceedings{DBLP:conf/cikm/0008ZL23,
  author    = {Heng Yang and Chen Zhang and Ke Li},
  title     = {PyABSA: A Modularized Framework for Reproducible Aspect-based Sentiment Analysis},
  booktitle = {Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, CIKM 2023},
  pages     = {5117--5122},
  publisher = {ACM},
  year      = {2023},
  doi       = {10.1145/3583780.3614752}
}

@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}
}