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: 14 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 |
---|---|
7.0 |
|
1.0 |
|
6.0 |
|
12.0 |
|
5.0 |
|
0.0 |
|
10.0 |
|
8.0 |
|
11.0 |
|
13.0 |
|
2.0 |
|
9.0 |
|
4.0 |
|
3.0 |
|
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_sl28")
# Run inference
preds = model("애몰라이트 후레쉬 AM1 표준슬립 손전등 스포츠/레저>캠핑>랜턴>손전등")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 7.8108 | 22 |
Label | Training Sample Count |
---|---|
0.0 | 70 |
1.0 | 70 |
2.0 | 70 |
3.0 | 25 |
4.0 | 30 |
5.0 | 70 |
6.0 | 70 |
7.0 | 70 |
8.0 | 70 |
9.0 | 70 |
10.0 | 70 |
11.0 | 70 |
12.0 | 70 |
13.0 | 26 |
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 |
---|---|---|---|
0.0060 | 1 | 0.5164 | - |
0.2994 | 50 | 0.4984 | - |
0.5988 | 100 | 0.4882 | - |
0.8982 | 150 | 0.1544 | - |
1.1976 | 200 | 0.0264 | - |
1.4970 | 250 | 0.0089 | - |
1.7964 | 300 | 0.0027 | - |
2.0958 | 350 | 0.0003 | - |
2.3952 | 400 | 0.0002 | - |
2.6946 | 450 | 0.0001 | - |
2.9940 | 500 | 0.0001 | - |
3.2934 | 550 | 0.0001 | - |
3.5928 | 600 | 0.0001 | - |
3.8922 | 650 | 0.0001 | - |
4.1916 | 700 | 0.0001 | - |
4.4910 | 750 | 0.0 | - |
4.7904 | 800 | 0.0 | - |
5.0898 | 850 | 0.0 | - |
5.3892 | 900 | 0.0 | - |
5.6886 | 950 | 0.0 | - |
5.9880 | 1000 | 0.0 | - |
6.2874 | 1050 | 0.0 | - |
6.5868 | 1100 | 0.0 | - |
6.8862 | 1150 | 0.0 | - |
7.1856 | 1200 | 0.0 | - |
7.4850 | 1250 | 0.0 | - |
7.7844 | 1300 | 0.0 | - |
8.0838 | 1350 | 0.0 | - |
8.3832 | 1400 | 0.0 | - |
8.6826 | 1450 | 0.0 | - |
8.9820 | 1500 | 0.0 | - |
9.2814 | 1550 | 0.0 | - |
9.5808 | 1600 | 0.0 | - |
9.8802 | 1650 | 0.0 | - |
10.1796 | 1700 | 0.0 | - |
10.4790 | 1750 | 0.0 | - |
10.7784 | 1800 | 0.0 | - |
11.0778 | 1850 | 0.0 | - |
11.3772 | 1900 | 0.0 | - |
11.6766 | 1950 | 0.0 | - |
11.9760 | 2000 | 0.0 | - |
12.2754 | 2050 | 0.0 | - |
12.5749 | 2100 | 0.0 | - |
12.8743 | 2150 | 0.0 | - |
13.1737 | 2200 | 0.0 | - |
13.4731 | 2250 | 0.0 | - |
13.7725 | 2300 | 0.0 | - |
14.0719 | 2350 | 0.0 | - |
14.3713 | 2400 | 0.0 | - |
14.6707 | 2450 | 0.0 | - |
14.9701 | 2500 | 0.0 | - |
15.2695 | 2550 | 0.0 | - |
15.5689 | 2600 | 0.0 | - |
15.8683 | 2650 | 0.0 | - |
16.1677 | 2700 | 0.0 | - |
16.4671 | 2750 | 0.0 | - |
16.7665 | 2800 | 0.0 | - |
17.0659 | 2850 | 0.0 | - |
17.3653 | 2900 | 0.0 | - |
17.6647 | 2950 | 0.0 | - |
17.9641 | 3000 | 0.0 | - |
18.2635 | 3050 | 0.0 | - |
18.5629 | 3100 | 0.0 | - |
18.8623 | 3150 | 0.0 | - |
19.1617 | 3200 | 0.0 | - |
19.4611 | 3250 | 0.0 | - |
19.7605 | 3300 | 0.0 | - |
20.0599 | 3350 | 0.0 | - |
20.3593 | 3400 | 0.0 | - |
20.6587 | 3450 | 0.0 | - |
20.9581 | 3500 | 0.0 | - |
21.2575 | 3550 | 0.0 | - |
21.5569 | 3600 | 0.0 | - |
21.8563 | 3650 | 0.0 | - |
22.1557 | 3700 | 0.0 | - |
22.4551 | 3750 | 0.0 | - |
22.7545 | 3800 | 0.0 | - |
23.0539 | 3850 | 0.0 | - |
23.3533 | 3900 | 0.0 | - |
23.6527 | 3950 | 0.0 | - |
23.9521 | 4000 | 0.0 | - |
24.2515 | 4050 | 0.0 | - |
24.5509 | 4100 | 0.0 | - |
24.8503 | 4150 | 0.0 | - |
25.1497 | 4200 | 0.0 | - |
25.4491 | 4250 | 0.0 | - |
25.7485 | 4300 | 0.0 | - |
26.0479 | 4350 | 0.0 | - |
26.3473 | 4400 | 0.0 | - |
26.6467 | 4450 | 0.0 | - |
26.9461 | 4500 | 0.0 | - |
27.2455 | 4550 | 0.0 | - |
27.5449 | 4600 | 0.0 | - |
27.8443 | 4650 | 0.0 | - |
28.1437 | 4700 | 0.0 | - |
28.4431 | 4750 | 0.0 | - |
28.7425 | 4800 | 0.0 | - |
29.0419 | 4850 | 0.0 | - |
29.3413 | 4900 | 0.0 | - |
29.6407 | 4950 | 0.0 | - |
29.9401 | 5000 | 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
- 157
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.