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 |
|
| 5.0 |
|
| 2.0 |
|
| 3.0 |
|
| 1.0 |
|
| 4.0 |
|
| 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_fi12")
# Run inference
preds = model("Q아라벨르 에린 극세사 침구세트 가구/인테리어>침구세트>이불베개세트>더블/퀸이불베개세트")
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 3 | 8.4207 | 20 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 48 |
| 1.0 | 7 |
| 2.0 | 70 |
| 3.0 | 70 |
| 4.0 | 6 |
| 5.0 | 70 |
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0189 | 1 | 0.4779 | - |
| 0.9434 | 50 | 0.4942 | - |
| 1.8868 | 100 | 0.3341 | - |
| 2.8302 | 150 | 0.012 | - |
| 3.7736 | 200 | 0.0006 | - |
| 4.7170 | 250 | 0.0005 | - |
| 5.6604 | 300 | 0.0007 | - |
| 6.6038 | 350 | 0.0005 | - |
| 7.5472 | 400 | 0.0007 | - |
| 8.4906 | 450 | 0.0004 | - |
| 9.4340 | 500 | 0.0004 | - |
| 10.3774 | 550 | 0.0 | - |
| 11.3208 | 600 | 0.0 | - |
| 12.2642 | 650 | 0.0 | - |
| 13.2075 | 700 | 0.0 | - |
| 14.1509 | 750 | 0.0 | - |
| 15.0943 | 800 | 0.0 | - |
| 16.0377 | 850 | 0.0 | - |
| 16.9811 | 900 | 0.0 | - |
| 17.9245 | 950 | 0.0 | - |
| 18.8679 | 1000 | 0.0 | - |
| 19.8113 | 1050 | 0.0 | - |
| 20.7547 | 1100 | 0.0 | - |
| 21.6981 | 1150 | 0.0 | - |
| 22.6415 | 1200 | 0.0 | - |
| 23.5849 | 1250 | 0.0 | - |
| 24.5283 | 1300 | 0.0 | - |
| 25.4717 | 1350 | 0.0 | - |
| 26.4151 | 1400 | 0.0 | - |
| 27.3585 | 1450 | 0.0 | - |
| 28.3019 | 1500 | 0.0 | - |
| 29.2453 | 1550 | 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}
}