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: 8 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 |
---|---|
1.0 |
|
5.0 |
|
2.0 |
|
4.0 |
|
0.0 |
|
3.0 |
|
7.0 |
|
6.0 |
|
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_sl30")
# Run inference
preds = model("알로 MATCH POINT 여성 테니스 스커트 스포츠/레저>테니스>테니스의류")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 8.2241 | 18 |
Label | Training Sample Count |
---|---|
0.0 | 70 |
1.0 | 70 |
2.0 | 70 |
3.0 | 70 |
4.0 | 50 |
5.0 | 70 |
6.0 | 70 |
7.0 | 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 |
---|---|---|---|
0.0094 | 1 | 0.4693 | - |
0.4717 | 50 | 0.4966 | - |
0.9434 | 100 | 0.2749 | - |
1.4151 | 150 | 0.0397 | - |
1.8868 | 200 | 0.0179 | - |
2.3585 | 250 | 0.0076 | - |
2.8302 | 300 | 0.0 | - |
3.3019 | 350 | 0.0 | - |
3.7736 | 400 | 0.0 | - |
4.2453 | 450 | 0.0 | - |
4.7170 | 500 | 0.0 | - |
5.1887 | 550 | 0.0 | - |
5.6604 | 600 | 0.0 | - |
6.1321 | 650 | 0.0 | - |
6.6038 | 700 | 0.0 | - |
7.0755 | 750 | 0.0 | - |
7.5472 | 800 | 0.0 | - |
8.0189 | 850 | 0.0 | - |
8.4906 | 900 | 0.0 | - |
8.9623 | 950 | 0.0 | - |
9.4340 | 1000 | 0.0 | - |
9.9057 | 1050 | 0.0 | - |
10.3774 | 1100 | 0.0 | - |
10.8491 | 1150 | 0.0 | - |
11.3208 | 1200 | 0.0 | - |
11.7925 | 1250 | 0.0 | - |
12.2642 | 1300 | 0.0 | - |
12.7358 | 1350 | 0.0 | - |
13.2075 | 1400 | 0.0 | - |
13.6792 | 1450 | 0.0 | - |
14.1509 | 1500 | 0.0 | - |
14.6226 | 1550 | 0.0 | - |
15.0943 | 1600 | 0.0 | - |
15.5660 | 1650 | 0.0 | - |
16.0377 | 1700 | 0.0 | - |
16.5094 | 1750 | 0.0 | - |
16.9811 | 1800 | 0.0 | - |
17.4528 | 1850 | 0.0 | - |
17.9245 | 1900 | 0.0 | - |
18.3962 | 1950 | 0.0 | - |
18.8679 | 2000 | 0.0 | - |
19.3396 | 2050 | 0.0 | - |
19.8113 | 2100 | 0.0 | - |
20.2830 | 2150 | 0.0 | - |
20.7547 | 2200 | 0.0 | - |
21.2264 | 2250 | 0.0 | - |
21.6981 | 2300 | 0.0 | - |
22.1698 | 2350 | 0.0 | - |
22.6415 | 2400 | 0.0 | - |
23.1132 | 2450 | 0.0 | - |
23.5849 | 2500 | 0.0 | - |
24.0566 | 2550 | 0.0 | - |
24.5283 | 2600 | 0.0 | - |
25.0 | 2650 | 0.0 | - |
25.4717 | 2700 | 0.0 | - |
25.9434 | 2750 | 0.0 | - |
26.4151 | 2800 | 0.0 | - |
26.8868 | 2850 | 0.0 | - |
27.3585 | 2900 | 0.0 | - |
27.8302 | 2950 | 0.0 | - |
28.3019 | 3000 | 0.0 | - |
28.7736 | 3050 | 0.0 | - |
29.2453 | 3100 | 0.0 | - |
29.7170 | 3150 | 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
- 100
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support