TwinTransitionMapper_AI

This repository contains the model for our paper entitled Not all twins are identical: the digital layer of “twin” transition market applications which is under review in Regional Studies (https://www.tandfonline.com/journals/cres20).

This is a SetFit model that can be used for Text Classification. This SetFit model uses intfloat/multilingual-e5-large as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

The model has been trained on paragraphs from German company websites 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.

The model is designed to predict the AI capabilities of German companies based on their website texts. It is intended to be used in conjunction with the [Twin_Transition_Mapper_Green model] (https://huggingface.co/LKriesch/TwinTransitionMapper_Green) to identify companies contributing to the twin transition in Germany. For detailed information on the fine-tuning process and the results of these models, please refer to the paper.

Model Description

• Model Type: SetFit
• Sentence Transformer body: intfloat/multilingual-e5-large
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 512 tokens

Model Sources

• Repository: SetFit on GitHub
• Paper: Efficient Few-Shot Learning Without Prompts
• Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("LKriesch/TwinTransitionMapper_AI")
# Run inference
preds = model("I loved the spiderman movie!")

Training Details

Framework Versions

• Python: 3.9.19
• SetFit: 1.0.3
• Sentence Transformers: 3.0.1
• Transformers: 4.44.0
• PyTorch: 2.4.0+cu124
• Datasets: 2.16.1
• Tokenizers: 0.19.1

Citation

BibTeX

@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}
}