metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 아기 무릎보호대 유아 돌 성장판 패드 스마일 무릎보호대 베이지 출산/육아 > 매트/안전용품 > 무릎보호대
- text: 함소아화장품 포포패치 아이편해 유칼립투스 오렌지 X 6개 출산/육아 > 매트/안전용품 > 모기밴드/퇴치용품
- text: 돗투돗 아기 무릎보호대 롤리팝 이중 걸음마 보조기 성장판 돌 유아 아이보리 베이비바니 출산/육아 > 매트/안전용품 > 무릎보호대
- text: 콘센트 안전커버 마개 안전캡 아기 멀티탭 안전덮개 실리콘 보호캡 출산/육아 > 매트/안전용품 > 콘센트안전커버
- text: 다이소 원터치 콘센트 안전 커버 4P 56873 출산/육아 > 매트/안전용품 > 콘센트안전커버
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: mini1013/master_domain
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: 10 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 |
|---|---|
| 2.0 |
|
| 9.0 |
|
| 4.0 |
|
| 1.0 |
|
| 7.0 |
|
| 5.0 |
|
| 6.0 |
|
| 0.0 |
|
| 3.0 |
|
| 8.0 |
|
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_bc3")
# Run inference
preds = model("다이소 원터치 콘센트 안전 커버 4P 56873 출산/육아 > 매트/안전용품 > 콘센트안전커버")
Training Details
Training Set Metrics
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 9 | 14.4541 | 34 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 16 |
| 1.0 | 20 |
| 2.0 | 20 |
| 3.0 | 20 |
| 4.0 | 20 |
| 5.0 | 20 |
| 6.0 | 20 |
| 7.0 | 20 |
| 8.0 | 20 |
| 9.0 | 20 |
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 |
|---|---|---|---|
| 0.0256 | 1 | 0.4765 | - |
| 1.2821 | 50 | 0.4502 | - |
| 2.5641 | 100 | 0.204 | - |
| 3.8462 | 150 | 0.061 | - |
| 5.1282 | 200 | 0.0263 | - |
| 6.4103 | 250 | 0.0101 | - |
| 7.6923 | 300 | 0.0003 | - |
| 8.9744 | 350 | 0.0001 | - |
| 10.2564 | 400 | 0.0001 | - |
| 11.5385 | 450 | 0.0001 | - |
| 12.8205 | 500 | 0.0001 | - |
| 14.1026 | 550 | 0.0001 | - |
| 15.3846 | 600 | 0.0 | - |
| 16.6667 | 650 | 0.0 | - |
| 17.9487 | 700 | 0.0 | - |
| 19.2308 | 750 | 0.0 | - |
| 20.5128 | 800 | 0.0 | - |
| 21.7949 | 850 | 0.0 | - |
| 23.0769 | 900 | 0.0 | - |
| 24.3590 | 950 | 0.0 | - |
| 25.6410 | 1000 | 0.0 | - |
| 26.9231 | 1050 | 0.0 | - |
| 28.2051 | 1100 | 0.0 | - |
| 29.4872 | 1150 | 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}
}