setfit-news-topic-paragraphs

joshuapsa/setfit-news-topic-paragraphs

This is a SetFit model that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:

1. Fine-tuning a Sentence Transformer with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.

This model was finetuned with the custom dataset joshuapsa/gpt-generated-news-paragraphs, which is a synthetic dataset containing news paragraphs and their topics.<br> Please refer to this to understand the label meanings of the prediction output.

Usage

To use this model for inference, first install the SetFit library:

python -m pip install setfit

You can then run inference as follows:

from setfit import SetFitModel

# Download from Hub and run inference
model = SetFitModel.from_pretrained("joshuapsa/setfit-news-topic-paragraphs")
# Run inference
preds = model(["Tensions escalated in the Taiwan Strait as Chinese and Taiwanese naval vessels engaged in a standoff, raising fears of a potential conflict. It has been 6 months since both sides started to engage in confrontational moves.",\
 "Cape Town, South Africa, grappled with social unrest as citizens protested against inadequate government response to a public health crisis. The demonstrations brought attention to the intersection of healthcare, governance, and societal well-being."])
# The underlying model body of the setfit model is a SentenceTransformer model, hence you can use it to encode a raw sentence into dense embeddings:
emb = model.model_body.encode("Your sentence goes here")

BibTeX entry and citation info

@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}