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 mini1013/master_domain 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 |
|---|---|
| 0.0 |
|
| 3.0 |
|
| 1.0 |
|
| 5.0 |
|
| 4.0 |
|
| 6.0 |
|
| 2.0 |
|
| Label | Accuracy |
|---|---|
| all | 0.7155 |
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("mini1013/master_cate_bt4_test")
# Run inference
preds = model("나스 래디언스 프라이머 30ml(SPF35) 옵션없음 블루밍컴퍼니")
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 5 | 9.7872 | 19 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 19 |
| 1.0 | 21 |
| 2.0 | 10 |
| 3.0 | 19 |
| 4.0 | 28 |
| 5.0 | 23 |
| 6.0 | 21 |
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0588 | 1 | 0.499 | - |
| 2.9412 | 50 | 0.3295 | - |
| 5.8824 | 100 | 0.0469 | - |
| 8.8235 | 150 | 0.0217 | - |
| 11.7647 | 200 | 0.0013 | - |
| 14.7059 | 250 | 0.0001 | - |
| 17.6471 | 300 | 0.0001 | - |
| 20.5882 | 350 | 0.0 | - |
| 23.5294 | 400 | 0.0 | - |
| 26.4706 | 450 | 0.0 | - |
| 29.4118 | 500 | 0.0 | - |
| 32.3529 | 550 | 0.0 | - |
| 35.2941 | 600 | 0.0 | - |
| 38.2353 | 650 | 0.0 | - |
| 41.1765 | 700 | 0.0 | - |
| 44.1176 | 750 | 0.0 | - |
| 47.0588 | 800 | 0.0 | - |
| 50.0 | 850 | 0.0 | - |
@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}
}