metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: >-
고흥 유자 10kg 신선농장 상품 10kg (#M)식품>농산물>과일>오렌지 T200 > Naverstore > 식품 > 과일 >
오렌지/자몽/레몬 > 오렌지
- text: >-
국내산 참다래 5kg/10kg 골드/레드키위 국내산 그린키위10kg 91-100과 (#M)식품>농산물>과일>키위/참다래 T200 >
Naverstore > 식품 > 과일 > 키위/참다래
- text: >-
제스프리 썬골드키위 대왕점보사이즈 8개입 (155g내외) 뉴질랜드 (#M)식품>농산물>과일>바나나 T200 > Naverstore
> 식품 > 과일 > 바나나/망고/파인애플 > 바나나
- text: >-
새콤달콤 스타루비 레드 자몽 특대과 특품 5kg 5-7과 (#M)식품>농산물>과일>자몽 T200 > Naverstore > 식품 >
과일 > 오렌지/자몽/레몬 > 자몽
- text: >-
프리미엄 냉동과일 아보카도 다이스, 냉동아보카도 500g 02. 냉동 람부탄 1팩 (#M)식품>농산물>과일>아보카도 T200 >
Naverstore > 식품 > 과일 > 아보카도
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: klue/roberta-base
model-index:
- name: SetFit with klue/roberta-base
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 0.998283105022831
name: Accuracy
SetFit with klue/roberta-base
This is a SetFit model that can be used for Text Classification. This SetFit model uses klue/roberta-base 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: klue/roberta-base
- Classification head: a LogisticRegression instance
- Maximum Sequence Length: 512 tokens
- Number of Classes: 31 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 |
|---|---|
| 20.0 |
|
| 19.0 |
|
| 27.0 |
|
| 8.0 |
|
| 17.0 |
|
| 0.0 |
|
| 21.0 |
|
| 4.0 |
|
| 18.0 |
|
| 13.0 |
|
| 22.0 |
|
| 3.0 |
|
| 14.0 |
|
| 24.0 |
|
| 23.0 |
|
| 2.0 |
|
| 9.0 |
|
| 11.0 |
|
| 30.0 |
|
| 15.0 |
|
| 7.0 |
|
| 1.0 |
|
| 10.0 |
|
| 28.0 |
|
| 6.0 |
|
| 25.0 |
|
| 26.0 |
|
| 5.0 |
|
| 29.0 |
|
| 12.0 |
|
| 16.0 |
|
Evaluation
Metrics
| Label | Accuracy |
|---|---|
| all | 0.9983 |
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_top_fd4")
# Run inference
preds = model("고흥 유자 10kg 신선농장 상품 10kg (#M)식품>농산물>과일>오렌지 T200 > Naverstore > 식품 > 과일 > 오렌지/자몽/레몬 > 오렌지")
Training Details
Training Set Metrics
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 14 | 21.8097 | 48 |
| Label | Training Sample Count |
|---|---|
| 0.0 | 50 |
| 1.0 | 50 |
| 2.0 | 50 |
| 3.0 | 50 |
| 4.0 | 50 |
| 5.0 | 50 |
| 6.0 | 50 |
| 7.0 | 50 |
| 8.0 | 50 |
| 9.0 | 50 |
| 10.0 | 50 |
| 11.0 | 50 |
| 12.0 | 50 |
| 13.0 | 50 |
| 14.0 | 50 |
| 15.0 | 50 |
| 16.0 | 50 |
| 17.0 | 50 |
| 18.0 | 50 |
| 19.0 | 50 |
| 20.0 | 50 |
| 21.0 | 50 |
| 22.0 | 50 |
| 23.0 | 50 |
| 24.0 | 50 |
| 25.0 | 50 |
| 26.0 | 50 |
| 27.0 | 50 |
| 28.0 | 50 |
| 29.0 | 50 |
| 30.0 | 50 |
Training Hyperparameters
- batch_size: (64, 64)
- num_epochs: (20, 20)
- max_steps: -1
- sampling_strategy: oversampling
- num_iterations: 30
- 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 |
|---|---|---|---|
| 0.0014 | 1 | 0.5541 | - |
| 0.0688 | 50 | 0.4773 | - |
| 0.1376 | 100 | 0.4113 | - |
| 0.2063 | 150 | 0.3227 | - |
| 0.2751 | 200 | 0.2635 | - |
| 0.3439 | 250 | 0.1443 | - |
| 0.4127 | 300 | 0.054 | - |
| 0.4814 | 350 | 0.031 | - |
| 0.5502 | 400 | 0.0242 | - |
| 0.6190 | 450 | 0.0186 | - |
| 0.6878 | 500 | 0.013 | - |
| 0.7565 | 550 | 0.0088 | - |
| 0.8253 | 600 | 0.0088 | - |
| 0.8941 | 650 | 0.0088 | - |
| 0.9629 | 700 | 0.0074 | - |
| 1.0316 | 750 | 0.0064 | - |
| 1.1004 | 800 | 0.0037 | - |
| 1.1692 | 850 | 0.0032 | - |
| 1.2380 | 900 | 0.002 | - |
| 1.3067 | 950 | 0.0018 | - |
| 1.3755 | 1000 | 0.0012 | - |
| 1.4443 | 1050 | 0.0006 | - |
| 1.5131 | 1100 | 0.0005 | - |
| 1.5818 | 1150 | 0.0004 | - |
| 1.6506 | 1200 | 0.0003 | - |
| 1.7194 | 1250 | 0.0003 | - |
| 1.7882 | 1300 | 0.0003 | - |
| 1.8569 | 1350 | 0.0003 | - |
| 1.9257 | 1400 | 0.0002 | - |
| 1.9945 | 1450 | 0.0002 | - |
| 2.0633 | 1500 | 0.0002 | - |
| 2.1320 | 1550 | 0.0002 | - |
| 2.2008 | 1600 | 0.0002 | - |
| 2.2696 | 1650 | 0.0001 | - |
| 2.3384 | 1700 | 0.0002 | - |
| 2.4072 | 1750 | 0.0001 | - |
| 2.4759 | 1800 | 0.0001 | - |
| 2.5447 | 1850 | 0.0001 | - |
| 2.6135 | 1900 | 0.0001 | - |
| 2.6823 | 1950 | 0.0001 | - |
| 2.7510 | 2000 | 0.0001 | - |
| 2.8198 | 2050 | 0.0001 | - |
| 2.8886 | 2100 | 0.0001 | - |
| 2.9574 | 2150 | 0.0001 | - |
| 3.0261 | 2200 | 0.0001 | - |
| 3.0949 | 2250 | 0.0001 | - |
| 3.1637 | 2300 | 0.0001 | - |
| 3.2325 | 2350 | 0.0001 | - |
| 3.3012 | 2400 | 0.0001 | - |
| 3.3700 | 2450 | 0.0001 | - |
| 3.4388 | 2500 | 0.0001 | - |
| 3.5076 | 2550 | 0.0001 | - |
| 3.5763 | 2600 | 0.0001 | - |
| 3.6451 | 2650 | 0.0001 | - |
| 3.7139 | 2700 | 0.0001 | - |
| 3.7827 | 2750 | 0.0001 | - |
| 3.8514 | 2800 | 0.0001 | - |
| 3.9202 | 2850 | 0.0001 | - |
| 3.9890 | 2900 | 0.0001 | - |
| 4.0578 | 2950 | 0.0001 | - |
| 4.1265 | 3000 | 0.0001 | - |
| 4.1953 | 3050 | 0.0001 | - |
| 4.2641 | 3100 | 0.0001 | - |
| 4.3329 | 3150 | 0.0001 | - |
| 4.4017 | 3200 | 0.0001 | - |
| 4.4704 | 3250 | 0.0001 | - |
| 4.5392 | 3300 | 0.0 | - |
| 4.6080 | 3350 | 0.0 | - |
| 4.6768 | 3400 | 0.0 | - |
| 4.7455 | 3450 | 0.0 | - |
| 4.8143 | 3500 | 0.0 | - |
| 4.8831 | 3550 | 0.0 | - |
| 4.9519 | 3600 | 0.0 | - |
| 5.0206 | 3650 | 0.0 | - |
| 5.0894 | 3700 | 0.0 | - |
| 5.1582 | 3750 | 0.0 | - |
| 5.2270 | 3800 | 0.0 | - |
| 5.2957 | 3850 | 0.0 | - |
| 5.3645 | 3900 | 0.0 | - |
| 5.4333 | 3950 | 0.0 | - |
| 5.5021 | 4000 | 0.0 | - |
| 5.5708 | 4050 | 0.0 | - |
| 5.6396 | 4100 | 0.0 | - |
| 5.7084 | 4150 | 0.0 | - |
| 5.7772 | 4200 | 0.0 | - |
| 5.8459 | 4250 | 0.0 | - |
| 5.9147 | 4300 | 0.0 | - |
| 5.9835 | 4350 | 0.0 | - |
| 6.0523 | 4400 | 0.0 | - |
| 6.1210 | 4450 | 0.0 | - |
| 6.1898 | 4500 | 0.0 | - |
| 6.2586 | 4550 | 0.0 | - |
| 6.3274 | 4600 | 0.0 | - |
| 6.3961 | 4650 | 0.0 | - |
| 6.4649 | 4700 | 0.0 | - |
| 6.5337 | 4750 | 0.0 | - |
| 6.6025 | 4800 | 0.0 | - |
| 6.6713 | 4850 | 0.0 | - |
| 6.7400 | 4900 | 0.0 | - |
| 6.8088 | 4950 | 0.0 | - |
| 6.8776 | 5000 | 0.0 | - |
| 6.9464 | 5050 | 0.0 | - |
| 7.0151 | 5100 | 0.0 | - |
| 7.0839 | 5150 | 0.0 | - |
| 7.1527 | 5200 | 0.0 | - |
| 7.2215 | 5250 | 0.0 | - |
| 7.2902 | 5300 | 0.0 | - |
| 7.3590 | 5350 | 0.0 | - |
| 7.4278 | 5400 | 0.0 | - |
| 7.4966 | 5450 | 0.0 | - |
| 7.5653 | 5500 | 0.0 | - |
| 7.6341 | 5550 | 0.0 | - |
| 7.7029 | 5600 | 0.0 | - |
| 7.7717 | 5650 | 0.0 | - |
| 7.8404 | 5700 | 0.0 | - |
| 7.9092 | 5750 | 0.0 | - |
| 7.9780 | 5800 | 0.0 | - |
| 8.0468 | 5850 | 0.0 | - |
| 8.1155 | 5900 | 0.0 | - |
| 8.1843 | 5950 | 0.0 | - |
| 8.2531 | 6000 | 0.0 | - |
| 8.3219 | 6050 | 0.0 | - |
| 8.3906 | 6100 | 0.0 | - |
| 8.4594 | 6150 | 0.0 | - |
| 8.5282 | 6200 | 0.0 | - |
| 8.5970 | 6250 | 0.0 | - |
| 8.6657 | 6300 | 0.0 | - |
| 8.7345 | 6350 | 0.0 | - |
| 8.8033 | 6400 | 0.0 | - |
| 8.8721 | 6450 | 0.0 | - |
| 8.9409 | 6500 | 0.0 | - |
| 9.0096 | 6550 | 0.0 | - |
| 9.0784 | 6600 | 0.0 | - |
| 9.1472 | 6650 | 0.0 | - |
| 9.2160 | 6700 | 0.0 | - |
| 9.2847 | 6750 | 0.0 | - |
| 9.3535 | 6800 | 0.0 | - |
| 9.4223 | 6850 | 0.0 | - |
| 9.4911 | 6900 | 0.0 | - |
| 9.5598 | 6950 | 0.0 | - |
| 9.6286 | 7000 | 0.0 | - |
| 9.6974 | 7050 | 0.0 | - |
| 9.7662 | 7100 | 0.0 | - |
| 9.8349 | 7150 | 0.0 | - |
| 9.9037 | 7200 | 0.0 | - |
| 9.9725 | 7250 | 0.0 | - |
| 10.0413 | 7300 | 0.0 | - |
| 10.1100 | 7350 | 0.0 | - |
| 10.1788 | 7400 | 0.0 | - |
| 10.2476 | 7450 | 0.0 | - |
| 10.3164 | 7500 | 0.0 | - |
| 10.3851 | 7550 | 0.0 | - |
| 10.4539 | 7600 | 0.0 | - |
| 10.5227 | 7650 | 0.0 | - |
| 10.5915 | 7700 | 0.0 | - |
| 10.6602 | 7750 | 0.0 | - |
| 10.7290 | 7800 | 0.0 | - |
| 10.7978 | 7850 | 0.0 | - |
| 10.8666 | 7900 | 0.0 | - |
| 10.9354 | 7950 | 0.0 | - |
| 11.0041 | 8000 | 0.0 | - |
| 11.0729 | 8050 | 0.0 | - |
| 11.1417 | 8100 | 0.0 | - |
| 11.2105 | 8150 | 0.0 | - |
| 11.2792 | 8200 | 0.0 | - |
| 11.3480 | 8250 | 0.0 | - |
| 11.4168 | 8300 | 0.0 | - |
| 11.4856 | 8350 | 0.0 | - |
| 11.5543 | 8400 | 0.0 | - |
| 11.6231 | 8450 | 0.0 | - |
| 11.6919 | 8500 | 0.0 | - |
| 11.7607 | 8550 | 0.0 | - |
| 11.8294 | 8600 | 0.0 | - |
| 11.8982 | 8650 | 0.0 | - |
| 11.9670 | 8700 | 0.0 | - |
| 12.0358 | 8750 | 0.0 | - |
| 12.1045 | 8800 | 0.0 | - |
| 12.1733 | 8850 | 0.0 | - |
| 12.2421 | 8900 | 0.0 | - |
| 12.3109 | 8950 | 0.0 | - |
| 12.3796 | 9000 | 0.0 | - |
| 12.4484 | 9050 | 0.0 | - |
| 12.5172 | 9100 | 0.0 | - |
| 12.5860 | 9150 | 0.0 | - |
| 12.6547 | 9200 | 0.0 | - |
| 12.7235 | 9250 | 0.0 | - |
| 12.7923 | 9300 | 0.0 | - |
| 12.8611 | 9350 | 0.0 | - |
| 12.9298 | 9400 | 0.0 | - |
| 12.9986 | 9450 | 0.0 | - |
| 13.0674 | 9500 | 0.0 | - |
| 13.1362 | 9550 | 0.0 | - |
| 13.2050 | 9600 | 0.0 | - |
| 13.2737 | 9650 | 0.0 | - |
| 13.3425 | 9700 | 0.0 | - |
| 13.4113 | 9750 | 0.0 | - |
| 13.4801 | 9800 | 0.0 | - |
| 13.5488 | 9850 | 0.0 | - |
| 13.6176 | 9900 | 0.0 | - |
| 13.6864 | 9950 | 0.0 | - |
| 13.7552 | 10000 | 0.0 | - |
| 13.8239 | 10050 | 0.0 | - |
| 13.8927 | 10100 | 0.0 | - |
| 13.9615 | 10150 | 0.0 | - |
| 14.0303 | 10200 | 0.0 | - |
| 14.0990 | 10250 | 0.0 | - |
| 14.1678 | 10300 | 0.0 | - |
| 14.2366 | 10350 | 0.0 | - |
| 14.3054 | 10400 | 0.0 | - |
| 14.3741 | 10450 | 0.0 | - |
| 14.4429 | 10500 | 0.0 | - |
| 14.5117 | 10550 | 0.0 | - |
| 14.5805 | 10600 | 0.0 | - |
| 14.6492 | 10650 | 0.0 | - |
| 14.7180 | 10700 | 0.0 | - |
| 14.7868 | 10750 | 0.0 | - |
| 14.8556 | 10800 | 0.0 | - |
| 14.9243 | 10850 | 0.0 | - |
| 14.9931 | 10900 | 0.0 | - |
| 15.0619 | 10950 | 0.0 | - |
| 15.1307 | 11000 | 0.0 | - |
| 15.1994 | 11050 | 0.0 | - |
| 15.2682 | 11100 | 0.0 | - |
| 15.3370 | 11150 | 0.0 | - |
| 15.4058 | 11200 | 0.0 | - |
| 15.4746 | 11250 | 0.0 | - |
| 15.5433 | 11300 | 0.0 | - |
| 15.6121 | 11350 | 0.0 | - |
| 15.6809 | 11400 | 0.0 | - |
| 15.7497 | 11450 | 0.0 | - |
| 15.8184 | 11500 | 0.0 | - |
| 15.8872 | 11550 | 0.0 | - |
| 15.9560 | 11600 | 0.0 | - |
| 16.0248 | 11650 | 0.0 | - |
| 16.0935 | 11700 | 0.0 | - |
| 16.1623 | 11750 | 0.0 | - |
| 16.2311 | 11800 | 0.0 | - |
| 16.2999 | 11850 | 0.0 | - |
| 16.3686 | 11900 | 0.0 | - |
| 16.4374 | 11950 | 0.0 | - |
| 16.5062 | 12000 | 0.0 | - |
| 16.5750 | 12050 | 0.0 | - |
| 16.6437 | 12100 | 0.0 | - |
| 16.7125 | 12150 | 0.0 | - |
| 16.7813 | 12200 | 0.0 | - |
| 16.8501 | 12250 | 0.0 | - |
| 16.9188 | 12300 | 0.0 | - |
| 16.9876 | 12350 | 0.0 | - |
| 17.0564 | 12400 | 0.0 | - |
| 17.1252 | 12450 | 0.0 | - |
| 17.1939 | 12500 | 0.0 | - |
| 17.2627 | 12550 | 0.0 | - |
| 17.3315 | 12600 | 0.0 | - |
| 17.4003 | 12650 | 0.0 | - |
| 17.4691 | 12700 | 0.0 | - |
| 17.5378 | 12750 | 0.0 | - |
| 17.6066 | 12800 | 0.0 | - |
| 17.6754 | 12850 | 0.0005 | - |
| 17.7442 | 12900 | 0.0021 | - |
| 17.8129 | 12950 | 0.0002 | - |
| 17.8817 | 13000 | 0.0001 | - |
| 17.9505 | 13050 | 0.0 | - |
| 18.0193 | 13100 | 0.0 | - |
| 18.0880 | 13150 | 0.0 | - |
| 18.1568 | 13200 | 0.0 | - |
| 18.2256 | 13250 | 0.0 | - |
| 18.2944 | 13300 | 0.0 | - |
| 18.3631 | 13350 | 0.0 | - |
| 18.4319 | 13400 | 0.0 | - |
| 18.5007 | 13450 | 0.0 | - |
| 18.5695 | 13500 | 0.0 | - |
| 18.6382 | 13550 | 0.0 | - |
| 18.7070 | 13600 | 0.0 | - |
| 18.7758 | 13650 | 0.0 | - |
| 18.8446 | 13700 | 0.0 | - |
| 18.9133 | 13750 | 0.0 | - |
| 18.9821 | 13800 | 0.0 | - |
| 19.0509 | 13850 | 0.0 | - |
| 19.1197 | 13900 | 0.0 | - |
| 19.1884 | 13950 | 0.0 | - |
| 19.2572 | 14000 | 0.0 | - |
| 19.3260 | 14050 | 0.0 | - |
| 19.3948 | 14100 | 0.0 | - |
| 19.4635 | 14150 | 0.0 | - |
| 19.5323 | 14200 | 0.0 | - |
| 19.6011 | 14250 | 0.0 | - |
| 19.6699 | 14300 | 0.0 | - |
| 19.7387 | 14350 | 0.0 | - |
| 19.8074 | 14400 | 0.0 | - |
| 19.8762 | 14450 | 0.0 | - |
| 19.9450 | 14500 | 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}
}