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: 6 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 |
|
0.0 |
|
5.0 |
|
4.0 |
|
3.0 |
|
1.0 |
|
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_fi10")
# Run inference
preds = model("미드센추리 투명 아크릴 스테인리스 트롤리 이동식 거실 테이블 가구/인테리어>주방가구>왜건/카트")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 9.0476 | 15 |
Label | Training Sample Count |
---|---|
0.0 | 70 |
1.0 | 70 |
2.0 | 70 |
3.0 | 70 |
4.0 | 70 |
5.0 | 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 |
---|---|---|---|
0.0120 | 1 | 0.494 | - |
0.6024 | 50 | 0.4972 | - |
1.2048 | 100 | 0.4906 | - |
1.8072 | 150 | 0.1734 | - |
2.4096 | 200 | 0.0195 | - |
3.0120 | 250 | 0.0002 | - |
3.6145 | 300 | 0.0 | - |
4.2169 | 350 | 0.0 | - |
4.8193 | 400 | 0.0001 | - |
5.4217 | 450 | 0.0 | - |
6.0241 | 500 | 0.0 | - |
6.6265 | 550 | 0.0 | - |
7.2289 | 600 | 0.0 | - |
7.8313 | 650 | 0.0 | - |
8.4337 | 700 | 0.0 | - |
9.0361 | 750 | 0.0 | - |
9.6386 | 800 | 0.0 | - |
10.2410 | 850 | 0.0 | - |
10.8434 | 900 | 0.0 | - |
11.4458 | 950 | 0.0 | - |
12.0482 | 1000 | 0.0 | - |
12.6506 | 1050 | 0.0 | - |
13.2530 | 1100 | 0.0 | - |
13.8554 | 1150 | 0.0 | - |
14.4578 | 1200 | 0.0 | - |
15.0602 | 1250 | 0.0 | - |
15.6627 | 1300 | 0.0 | - |
16.2651 | 1350 | 0.0 | - |
16.8675 | 1400 | 0.0 | - |
17.4699 | 1450 | 0.0 | - |
18.0723 | 1500 | 0.0 | - |
18.6747 | 1550 | 0.0 | - |
19.2771 | 1600 | 0.0 | - |
19.8795 | 1650 | 0.0 | - |
20.4819 | 1700 | 0.0 | - |
21.0843 | 1750 | 0.0 | - |
21.6867 | 1800 | 0.0 | - |
22.2892 | 1850 | 0.0 | - |
22.8916 | 1900 | 0.0 | - |
23.4940 | 1950 | 0.0 | - |
24.0964 | 2000 | 0.0 | - |
24.6988 | 2050 | 0.0 | - |
25.3012 | 2100 | 0.0 | - |
25.9036 | 2150 | 0.0 | - |
26.5060 | 2200 | 0.0 | - |
27.1084 | 2250 | 0.0 | - |
27.7108 | 2300 | 0.0 | - |
28.3133 | 2350 | 0.0 | - |
28.9157 | 2400 | 0.0 | - |
29.5181 | 2450 | 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}
}
- Downloads last month
- 1,184
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.