metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 머리감는의자 샴푸베드 샴푸대 가정용 목욕침대 세안기 어린이 아기 접의식 블루 출산/육아 > 목욕용품 > 기타목욕용품
- text: >-
여성 목욕 유아 샤워 웨딩 플라워 타월 타올 드레스 어린이 파티 가운 플레이 솔리드 잠옷 11=CM11_8-9T 130-140cm
출산/육아 > 목욕용품 > 유아목욕가운
- text: 아동용 레이어드나시반팔티 J4385 나시티11호 트임나시13호 출산/육아 > 목욕용품 > 유아목욕가운
- text: 욕실 타일 바닥 미끄럼방지 스티커 12P 세트 출산/육아 > 목욕용품 > 기타목욕용품
- text: 가정용 테이블 디지털 온습도 전자기계 욕조온도계 측정기 출산/육아 > 목욕용품 > 유아욕탕온도계
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: mini1013/master_domain
model-index:
- name: SetFit with mini1013/master_domain
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 1
name: Accuracy
SetFit with mini1013/master_domain
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:
- Fine-tuning a Sentence Transformer with contrastive learning.
- Training a classification head with features from the fine-tuned Sentence Transformer.
Model Details
Model Description
- Model Type: SetFit
- Sentence Transformer body: mini1013/master_domain
- Classification head: a LogisticRegression instance
- Maximum Sequence Length: 512 tokens
- Number of Classes: 11 classes
Model Sources
- Repository: SetFit on GitHub
- Paper: Efficient Few-Shot Learning Without Prompts
- Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts
Model Labels
| Label | Examples |
|---|---|
| 5.0 |
|
| 7.0 |
|
| 1.0 |
|
| 9.0 |
|
| 6.0 |
|
| 4.0 |
|
| 8.0 |
|
| 2.0 |
|
| 3.0 |
|
| 0.0 |
|
| 10.0 |
|
Evaluation
Metrics
| Label | Accuracy |
|---|---|
| all | 1.0 |
Uses
Direct Use for Inference
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_bc4")
# Run inference
preds = model("욕실 타일 바닥 미끄럼방지 스티커 12P 세트 출산/육아 > 목욕용품 > 기타목욕용품")
Training Details
Training Set Metrics
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 7 | 14.2403 | 27 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 70 |
| 1.0 | 70 |
| 2.0 | 70 |
| 3.0 | 70 |
| 4.0 | 70 |
| 5.0 | 70 |
| 6.0 | 70 |
| 7.0 | 70 |
| 8.0 | 70 |
| 9.0 | 70 |
| 10.0 | 70 |
Training Hyperparameters
- batch_size: (256, 256)
- num_epochs: (30, 30)
- max_steps: -1
- sampling_strategy: oversampling
- num_iterations: 50
- body_learning_rate: (2e-05, 1e-05)
- head_learning_rate: 0.01
- loss: CosineSimilarityLoss
- distance_metric: cosine_distance
- margin: 0.25
- end_to_end: False
- use_amp: False
- warmup_proportion: 0.1
- l2_weight: 0.01
- seed: 42
- eval_max_steps: -1
- load_best_model_at_end: False
Training Results
| Epoch | Step | Training Loss | Validation Loss |
|---|---|---|---|
| 0.0066 | 1 | 0.4891 | - |
| 0.3311 | 50 | 0.5008 | - |
| 0.6623 | 100 | 0.4057 | - |
| 0.9934 | 150 | 0.3132 | - |
| 1.3245 | 200 | 0.176 | - |
| 1.6556 | 250 | 0.0868 | - |
| 1.9868 | 300 | 0.0349 | - |
| 2.3179 | 350 | 0.0133 | - |
| 2.6490 | 400 | 0.0018 | - |
| 2.9801 | 450 | 0.0006 | - |
| 3.3113 | 500 | 0.0004 | - |
| 3.6424 | 550 | 0.0005 | - |
| 3.9735 | 600 | 0.0003 | - |
| 4.3046 | 650 | 0.0002 | - |
| 4.6358 | 700 | 0.0002 | - |
| 4.9669 | 750 | 0.0002 | - |
| 5.2980 | 800 | 0.0001 | - |
| 5.6291 | 850 | 0.0001 | - |
| 5.9603 | 900 | 0.0001 | - |
| 6.2914 | 950 | 0.0001 | - |
| 6.6225 | 1000 | 0.0001 | - |
| 6.9536 | 1050 | 0.0001 | - |
| 7.2848 | 1100 | 0.0001 | - |
| 7.6159 | 1150 | 0.0001 | - |
| 7.9470 | 1200 | 0.0001 | - |
| 8.2781 | 1250 | 0.0001 | - |
| 8.6093 | 1300 | 0.0001 | - |
| 8.9404 | 1350 | 0.0001 | - |
| 9.2715 | 1400 | 0.0001 | - |
| 9.6026 | 1450 | 0.0 | - |
| 9.9338 | 1500 | 0.0001 | - |
| 10.2649 | 1550 | 0.0 | - |
| 10.5960 | 1600 | 0.0 | - |
| 10.9272 | 1650 | 0.0 | - |
| 11.2583 | 1700 | 0.0 | - |
| 11.5894 | 1750 | 0.0 | - |
| 11.9205 | 1800 | 0.0 | - |
| 12.2517 | 1850 | 0.0 | - |
| 12.5828 | 1900 | 0.0 | - |
| 12.9139 | 1950 | 0.0 | - |
| 13.2450 | 2000 | 0.0 | - |
| 13.5762 | 2050 | 0.0 | - |
| 13.9073 | 2100 | 0.0 | - |
| 14.2384 | 2150 | 0.0 | - |
| 14.5695 | 2200 | 0.0 | - |
| 14.9007 | 2250 | 0.0 | - |
| 15.2318 | 2300 | 0.0 | - |
| 15.5629 | 2350 | 0.0 | - |
| 15.8940 | 2400 | 0.0 | - |
| 16.2252 | 2450 | 0.0 | - |
| 16.5563 | 2500 | 0.0 | - |
| 16.8874 | 2550 | 0.0 | - |
| 17.2185 | 2600 | 0.0 | - |
| 17.5497 | 2650 | 0.0 | - |
| 17.8808 | 2700 | 0.0 | - |
| 18.2119 | 2750 | 0.0 | - |
| 18.5430 | 2800 | 0.0 | - |
| 18.8742 | 2850 | 0.0 | - |
| 19.2053 | 2900 | 0.0 | - |
| 19.5364 | 2950 | 0.0 | - |
| 19.8675 | 3000 | 0.0 | - |
| 20.1987 | 3050 | 0.0 | - |
| 20.5298 | 3100 | 0.0 | - |
| 20.8609 | 3150 | 0.0 | - |
| 21.1921 | 3200 | 0.0 | - |
| 21.5232 | 3250 | 0.0 | - |
| 21.8543 | 3300 | 0.0 | - |
| 22.1854 | 3350 | 0.0 | - |
| 22.5166 | 3400 | 0.0 | - |
| 22.8477 | 3450 | 0.0 | - |
| 23.1788 | 3500 | 0.0 | - |
| 23.5099 | 3550 | 0.0 | - |
| 23.8411 | 3600 | 0.0 | - |
| 24.1722 | 3650 | 0.0 | - |
| 24.5033 | 3700 | 0.0 | - |
| 24.8344 | 3750 | 0.0 | - |
| 25.1656 | 3800 | 0.0 | - |
| 25.4967 | 3850 | 0.0 | - |
| 25.8278 | 3900 | 0.0 | - |
| 26.1589 | 3950 | 0.0 | - |
| 26.4901 | 4000 | 0.0 | - |
| 26.8212 | 4050 | 0.0 | - |
| 27.1523 | 4100 | 0.0 | - |
| 27.4834 | 4150 | 0.0 | - |
| 27.8146 | 4200 | 0.0 | - |
| 28.1457 | 4250 | 0.0 | - |
| 28.4768 | 4300 | 0.0 | - |
| 28.8079 | 4350 | 0.0 | - |
| 29.1391 | 4400 | 0.0 | - |
| 29.4702 | 4450 | 0.0 | - |
| 29.8013 | 4500 | 0.0 | - |
Framework Versions
- Python: 3.10.12
- SetFit: 1.1.0
- Sentence Transformers: 3.3.1
- Transformers: 4.44.2
- PyTorch: 2.2.0a0+81ea7a4
- Datasets: 3.2.0
- Tokenizers: 0.19.1
Citation
BibTeX
@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}
}