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: 18 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 |
---|---|
11.0 |
|
13.0 |
|
10.0 |
|
1.0 |
|
15.0 |
|
17.0 |
|
3.0 |
|
14.0 |
|
4.0 |
|
6.0 |
|
8.0 |
|
9.0 |
|
2.0 |
|
7.0 |
|
12.0 |
|
5.0 |
|
16.0 |
|
0.0 |
|
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_sl31")
# Run inference
preds = model("허리 단련 운동 허리강화 로마의자 로만체어 옆구리 스포츠/레저>헬스>복근운동기구")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 2 | 8.0378 | 18 |
Label | Training Sample Count |
---|---|
0.0 | 3 |
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 |
11.0 | 70 |
12.0 | 69 |
13.0 | 70 |
14.0 | 68 |
15.0 | 70 |
16.0 | 70 |
17.0 | 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 |
---|---|---|---|
0.0043 | 1 | 0.499 | - |
0.2146 | 50 | 0.4998 | - |
0.4292 | 100 | 0.4521 | - |
0.6438 | 150 | 0.2435 | - |
0.8584 | 200 | 0.093 | - |
1.0730 | 250 | 0.0291 | - |
1.2876 | 300 | 0.012 | - |
1.5021 | 350 | 0.0065 | - |
1.7167 | 400 | 0.0045 | - |
1.9313 | 450 | 0.0039 | - |
2.1459 | 500 | 0.0041 | - |
2.3605 | 550 | 0.0021 | - |
2.5751 | 600 | 0.0002 | - |
2.7897 | 650 | 0.0001 | - |
3.0043 | 700 | 0.0001 | - |
3.2189 | 750 | 0.0001 | - |
3.4335 | 800 | 0.0001 | - |
3.6481 | 850 | 0.0001 | - |
3.8627 | 900 | 0.0001 | - |
4.0773 | 950 | 0.0001 | - |
4.2918 | 1000 | 0.0001 | - |
4.5064 | 1050 | 0.0001 | - |
4.7210 | 1100 | 0.0001 | - |
4.9356 | 1150 | 0.0 | - |
5.1502 | 1200 | 0.0 | - |
5.3648 | 1250 | 0.0 | - |
5.5794 | 1300 | 0.0 | - |
5.7940 | 1350 | 0.0 | - |
6.0086 | 1400 | 0.0 | - |
6.2232 | 1450 | 0.0 | - |
6.4378 | 1500 | 0.0 | - |
6.6524 | 1550 | 0.0 | - |
6.8670 | 1600 | 0.0 | - |
7.0815 | 1650 | 0.0 | - |
7.2961 | 1700 | 0.0 | - |
7.5107 | 1750 | 0.0 | - |
7.7253 | 1800 | 0.0 | - |
7.9399 | 1850 | 0.0 | - |
8.1545 | 1900 | 0.0 | - |
8.3691 | 1950 | 0.0 | - |
8.5837 | 2000 | 0.0 | - |
8.7983 | 2050 | 0.0 | - |
9.0129 | 2100 | 0.0 | - |
9.2275 | 2150 | 0.0 | - |
9.4421 | 2200 | 0.0 | - |
9.6567 | 2250 | 0.0 | - |
9.8712 | 2300 | 0.0 | - |
10.0858 | 2350 | 0.0 | - |
10.3004 | 2400 | 0.0 | - |
10.5150 | 2450 | 0.0 | - |
10.7296 | 2500 | 0.0 | - |
10.9442 | 2550 | 0.0 | - |
11.1588 | 2600 | 0.0 | - |
11.3734 | 2650 | 0.0 | - |
11.5880 | 2700 | 0.0 | - |
11.8026 | 2750 | 0.0 | - |
12.0172 | 2800 | 0.0 | - |
12.2318 | 2850 | 0.0 | - |
12.4464 | 2900 | 0.0 | - |
12.6609 | 2950 | 0.0 | - |
12.8755 | 3000 | 0.0 | - |
13.0901 | 3050 | 0.0 | - |
13.3047 | 3100 | 0.0 | - |
13.5193 | 3150 | 0.0 | - |
13.7339 | 3200 | 0.0 | - |
13.9485 | 3250 | 0.0 | - |
14.1631 | 3300 | 0.0 | - |
14.3777 | 3350 | 0.0 | - |
14.5923 | 3400 | 0.0 | - |
14.8069 | 3450 | 0.0 | - |
15.0215 | 3500 | 0.0 | - |
15.2361 | 3550 | 0.0 | - |
15.4506 | 3600 | 0.0 | - |
15.6652 | 3650 | 0.0 | - |
15.8798 | 3700 | 0.0 | - |
16.0944 | 3750 | 0.0 | - |
16.3090 | 3800 | 0.0 | - |
16.5236 | 3850 | 0.0 | - |
16.7382 | 3900 | 0.0 | - |
16.9528 | 3950 | 0.0 | - |
17.1674 | 4000 | 0.0 | - |
17.3820 | 4050 | 0.0 | - |
17.5966 | 4100 | 0.0 | - |
17.8112 | 4150 | 0.0 | - |
18.0258 | 4200 | 0.0 | - |
18.2403 | 4250 | 0.0 | - |
18.4549 | 4300 | 0.0 | - |
18.6695 | 4350 | 0.0 | - |
18.8841 | 4400 | 0.0 | - |
19.0987 | 4450 | 0.0 | - |
19.3133 | 4500 | 0.0 | - |
19.5279 | 4550 | 0.0 | - |
19.7425 | 4600 | 0.0 | - |
19.9571 | 4650 | 0.0 | - |
20.1717 | 4700 | 0.0 | - |
20.3863 | 4750 | 0.0 | - |
20.6009 | 4800 | 0.0 | - |
20.8155 | 4850 | 0.0 | - |
21.0300 | 4900 | 0.0 | - |
21.2446 | 4950 | 0.0 | - |
21.4592 | 5000 | 0.0 | - |
21.6738 | 5050 | 0.0 | - |
21.8884 | 5100 | 0.0 | - |
22.1030 | 5150 | 0.0 | - |
22.3176 | 5200 | 0.0 | - |
22.5322 | 5250 | 0.0 | - |
22.7468 | 5300 | 0.0 | - |
22.9614 | 5350 | 0.0 | - |
23.1760 | 5400 | 0.0 | - |
23.3906 | 5450 | 0.0 | - |
23.6052 | 5500 | 0.0 | - |
23.8197 | 5550 | 0.0 | - |
24.0343 | 5600 | 0.0 | - |
24.2489 | 5650 | 0.0 | - |
24.4635 | 5700 | 0.0 | - |
24.6781 | 5750 | 0.0 | - |
24.8927 | 5800 | 0.0 | - |
25.1073 | 5850 | 0.0 | - |
25.3219 | 5900 | 0.0 | - |
25.5365 | 5950 | 0.0 | - |
25.7511 | 6000 | 0.0 | - |
25.9657 | 6050 | 0.0 | - |
26.1803 | 6100 | 0.0 | - |
26.3948 | 6150 | 0.0 | - |
26.6094 | 6200 | 0.0 | - |
26.8240 | 6250 | 0.0 | - |
27.0386 | 6300 | 0.0 | - |
27.2532 | 6350 | 0.0 | - |
27.4678 | 6400 | 0.0 | - |
27.6824 | 6450 | 0.0 | - |
27.8970 | 6500 | 0.0 | - |
28.1116 | 6550 | 0.0 | - |
28.3262 | 6600 | 0.0 | - |
28.5408 | 6650 | 0.0 | - |
28.7554 | 6700 | 0.0 | - |
28.9700 | 6750 | 0.0 | - |
29.1845 | 6800 | 0.0 | - |
29.3991 | 6850 | 0.0 | - |
29.6137 | 6900 | 0.0 | - |
29.8283 | 6950 | 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
- 119
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.