Efficient Few-Shot Learning Without Prompts
Paper
•
2209.11055
•
Published
•
4
This is a SetFit model that can be used for Text Classification. This SetFit model uses deutsche-telekom/gbert-large-paraphrase-cosine as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
| Label | Examples |
|---|---|
| 0 |
|
| 1 |
|
| Label | F1 | Precision | Recall |
|---|---|---|---|
| all | 0.8323 | 0.8360 | 0.8316 |
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("GeorgHCundK/gbert-large-stance-socialpolicy")
# Run inference
preds = model("Daran wollen wir anknüpfen und die Teilhabeleistungen stetig weiterentwickeln.")
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 2 | 15.1124 | 50 |
| Label | Training Sample Count |
|---|---|
| 0 | 392 |
| 1 | 364 |
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0002 | 1 | 0.3708 | - |
| 0.0112 | 50 | 0.2974 | 0.2649 |
| 0.0223 | 100 | 0.2567 | 0.2522 |
| 0.0335 | 150 | 0.2392 | 0.2338 |
| 0.0446 | 200 | 0.1895 | 0.1996 |
| 0.0558 | 250 | 0.0703 | 0.1907 |
| 0.0669 | 300 | 0.0126 | 0.2218 |
| 0.0781 | 350 | 0.0026 | 0.1964 |
| 0.0892 | 400 | 0.0009 | 0.2404 |
@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}
}
Base model
deepset/gbert-large