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 |
|---|---|
| 4.0 |
|
| 1.0 |
|
| 0.0 |
|
| 3.0 |
|
| 2.0 |
|
| Label | Accuracy |
|---|---|
| all | 1.0 |
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_sl26")
# Run inference
preds = model("스타스포츠 스타 루카스 스포츠용품 운동신발 족구화 스포츠/레저>족구>족구화")
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 2 | 8.0441 | 19 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 70 |
| 1.0 | 70 |
| 2.0 | 15 |
| 3.0 | 70 |
| 4.0 | 70 |
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0172 | 1 | 0.4882 | - |
| 0.8621 | 50 | 0.4668 | - |
| 1.7241 | 100 | 0.1034 | - |
| 2.5862 | 150 | 0.0002 | - |
| 3.4483 | 200 | 0.0 | - |
| 4.3103 | 250 | 0.0 | - |
| 5.1724 | 300 | 0.0 | - |
| 6.0345 | 350 | 0.0 | - |
| 6.8966 | 400 | 0.0 | - |
| 7.7586 | 450 | 0.0 | - |
| 8.6207 | 500 | 0.0 | - |
| 9.4828 | 550 | 0.0 | - |
| 10.3448 | 600 | 0.0 | - |
| 11.2069 | 650 | 0.0 | - |
| 12.0690 | 700 | 0.0 | - |
| 12.9310 | 750 | 0.0 | - |
| 13.7931 | 800 | 0.0 | - |
| 14.6552 | 850 | 0.0 | - |
| 15.5172 | 900 | 0.0 | - |
| 16.3793 | 950 | 0.0 | - |
| 17.2414 | 1000 | 0.0 | - |
| 18.1034 | 1050 | 0.0 | - |
| 18.9655 | 1100 | 0.0 | - |
| 19.8276 | 1150 | 0.0 | - |
| 20.6897 | 1200 | 0.0 | - |
| 21.5517 | 1250 | 0.0 | - |
| 22.4138 | 1300 | 0.0 | - |
| 23.2759 | 1350 | 0.0 | - |
| 24.1379 | 1400 | 0.0 | - |
| 25.0 | 1450 | 0.0 | - |
| 25.8621 | 1500 | 0.0 | - |
| 26.7241 | 1550 | 0.0 | - |
| 27.5862 | 1600 | 0.0 | - |
| 28.4483 | 1650 | 0.0 | - |
| 29.3103 | 1700 | 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}
}