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 |
|---|---|
| 5.0 |
|
| 13.0 |
|
| 1.0 |
|
| 11.0 |
|
| 2.0 |
|
| 3.0 |
|
| 6.0 |
|
| 7.0 |
|
| 0.0 |
|
| 10.0 |
|
| 8.0 |
|
| 4.0 |
|
| 12.0 |
|
| 9.0 |
|
| 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_fi7")
# Run inference
preds = model("오운 어린이 침대 프레임 SS 가구/인테리어>아동/주니어가구>침대>일반침대")
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 3 | 8.3744 | 18 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 65 |
| 1.0 | 70 |
| 2.0 | 70 |
| 3.0 | 70 |
| 4.0 | 37 |
| 5.0 | 70 |
| 6.0 | 21 |
| 7.0 | 70 |
| 8.0 | 70 |
| 9.0 | 70 |
| 10.0 | 70 |
| 11.0 | 70 |
| 12.0 | 70 |
| 13.0 | 69 |
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0057 | 1 | 0.5123 | - |
| 0.2857 | 50 | 0.5012 | - |
| 0.5714 | 100 | 0.3699 | - |
| 0.8571 | 150 | 0.1028 | - |
| 1.1429 | 200 | 0.0304 | - |
| 1.4286 | 250 | 0.0147 | - |
| 1.7143 | 300 | 0.012 | - |
| 2.0 | 350 | 0.009 | - |
| 2.2857 | 400 | 0.0074 | - |
| 2.5714 | 450 | 0.0033 | - |
| 2.8571 | 500 | 0.004 | - |
| 3.1429 | 550 | 0.0036 | - |
| 3.4286 | 600 | 0.0036 | - |
| 3.7143 | 650 | 0.0036 | - |
| 4.0 | 700 | 0.0036 | - |
| 4.2857 | 750 | 0.0027 | - |
| 4.5714 | 800 | 0.0034 | - |
| 4.8571 | 850 | 0.004 | - |
| 5.1429 | 900 | 0.0016 | - |
| 5.4286 | 950 | 0.0001 | - |
| 5.7143 | 1000 | 0.0 | - |
| 6.0 | 1050 | 0.0 | - |
| 6.2857 | 1100 | 0.0 | - |
| 6.5714 | 1150 | 0.0 | - |
| 6.8571 | 1200 | 0.0 | - |
| 7.1429 | 1250 | 0.0 | - |
| 7.4286 | 1300 | 0.0 | - |
| 7.7143 | 1350 | 0.0 | - |
| 8.0 | 1400 | 0.0 | - |
| 8.2857 | 1450 | 0.0 | - |
| 8.5714 | 1500 | 0.0 | - |
| 8.8571 | 1550 | 0.0 | - |
| 9.1429 | 1600 | 0.0 | - |
| 9.4286 | 1650 | 0.0 | - |
| 9.7143 | 1700 | 0.0 | - |
| 10.0 | 1750 | 0.0 | - |
| 10.2857 | 1800 | 0.0 | - |
| 10.5714 | 1850 | 0.0 | - |
| 10.8571 | 1900 | 0.0 | - |
| 11.1429 | 1950 | 0.0 | - |
| 11.4286 | 2000 | 0.0 | - |
| 11.7143 | 2050 | 0.0 | - |
| 12.0 | 2100 | 0.0 | - |
| 12.2857 | 2150 | 0.0 | - |
| 12.5714 | 2200 | 0.0 | - |
| 12.8571 | 2250 | 0.0 | - |
| 13.1429 | 2300 | 0.0 | - |
| 13.4286 | 2350 | 0.0 | - |
| 13.7143 | 2400 | 0.0 | - |
| 14.0 | 2450 | 0.0 | - |
| 14.2857 | 2500 | 0.0 | - |
| 14.5714 | 2550 | 0.0 | - |
| 14.8571 | 2600 | 0.0 | - |
| 15.1429 | 2650 | 0.0 | - |
| 15.4286 | 2700 | 0.0 | - |
| 15.7143 | 2750 | 0.0 | - |
| 16.0 | 2800 | 0.0 | - |
| 16.2857 | 2850 | 0.0 | - |
| 16.5714 | 2900 | 0.0 | - |
| 16.8571 | 2950 | 0.0 | - |
| 17.1429 | 3000 | 0.0 | - |
| 17.4286 | 3050 | 0.0 | - |
| 17.7143 | 3100 | 0.0 | - |
| 18.0 | 3150 | 0.0 | - |
| 18.2857 | 3200 | 0.0 | - |
| 18.5714 | 3250 | 0.0 | - |
| 18.8571 | 3300 | 0.0 | - |
| 19.1429 | 3350 | 0.0 | - |
| 19.4286 | 3400 | 0.0 | - |
| 19.7143 | 3450 | 0.0 | - |
| 20.0 | 3500 | 0.0 | - |
| 20.2857 | 3550 | 0.0 | - |
| 20.5714 | 3600 | 0.0 | - |
| 20.8571 | 3650 | 0.0 | - |
| 21.1429 | 3700 | 0.0 | - |
| 21.4286 | 3750 | 0.0 | - |
| 21.7143 | 3800 | 0.0 | - |
| 22.0 | 3850 | 0.0 | - |
| 22.2857 | 3900 | 0.0 | - |
| 22.5714 | 3950 | 0.0 | - |
| 22.8571 | 4000 | 0.0 | - |
| 23.1429 | 4050 | 0.0 | - |
| 23.4286 | 4100 | 0.0 | - |
| 23.7143 | 4150 | 0.0 | - |
| 24.0 | 4200 | 0.0 | - |
| 24.2857 | 4250 | 0.0 | - |
| 24.5714 | 4300 | 0.0 | - |
| 24.8571 | 4350 | 0.0 | - |
| 25.1429 | 4400 | 0.0 | - |
| 25.4286 | 4450 | 0.0 | - |
| 25.7143 | 4500 | 0.0 | - |
| 26.0 | 4550 | 0.0 | - |
| 26.2857 | 4600 | 0.0 | - |
| 26.5714 | 4650 | 0.0 | - |
| 26.8571 | 4700 | 0.0 | - |
| 27.1429 | 4750 | 0.0 | - |
| 27.4286 | 4800 | 0.0 | - |
| 27.7143 | 4850 | 0.0 | - |
| 28.0 | 4900 | 0.0 | - |
| 28.2857 | 4950 | 0.0 | - |
| 28.5714 | 5000 | 0.0 | - |
| 28.8571 | 5050 | 0.0 | - |
| 29.1429 | 5100 | 0.0 | - |
| 29.4286 | 5150 | 0.0 | - |
| 29.7143 | 5200 | 0.0 | - |
| 30.0 | 5250 | 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}
}