metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 바운티풀 프리미엄 코마사 사틴면 호텔 이불커버 Q 가구/인테리어>침구단품>이불커버
- text: 쇼파커버 사계절 담요 블랭킷 캠핑 이불 차박 대형 러그 가구/인테리어>침구단품>담요
- text: 플로라 시어서커 리플 여름 홑이불 SS 가구/인테리어>침구단품>홑이불
- text: 아이리스 포르토MT 모달 워싱 스프레드 Q 가구/인테리어>침구단품>스프레드
- text: 모던하우스 마이호텔 여름 모달혼방 고밀도워싱 차렵이불 S 가구/인테리어>침구단품>차렵이불
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: mini1013/master_domain
model-index:
- name: SetFit with mini1013/master_domain
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 1
name: Accuracy
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: 13 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 |
---|---|
0.0 |
|
9.0 |
|
10.0 |
|
11.0 |
|
1.0 |
|
2.0 |
|
8.0 |
|
3.0 |
|
4.0 |
|
7.0 |
|
5.0 |
|
12.0 |
|
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_fi11")
# Run inference
preds = model("플로라 시어서커 리플 여름 홑이불 SS 가구/인테리어>침구단품>홑이불")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 8.8067 | 23 |
Label | Training Sample Count |
---|---|
0.0 | 70 |
1.0 | 70 |
2.0 | 70 |
3.0 | 70 |
4.0 | 70 |
5.0 | 50 |
6.0 | 70 |
7.0 | 70 |
8.0 | 70 |
9.0 | 70 |
10.0 | 70 |
11.0 | 70 |
12.0 | 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 |
---|---|---|---|
0.0057 | 1 | 0.5104 | - |
0.2874 | 50 | 0.4986 | - |
0.5747 | 100 | 0.3956 | - |
0.8621 | 150 | 0.1871 | - |
1.1494 | 200 | 0.0555 | - |
1.4368 | 250 | 0.017 | - |
1.7241 | 300 | 0.0073 | - |
2.0115 | 350 | 0.0015 | - |
2.2989 | 400 | 0.0003 | - |
2.5862 | 450 | 0.0002 | - |
2.8736 | 500 | 0.0001 | - |
3.1609 | 550 | 0.0001 | - |
3.4483 | 600 | 0.0001 | - |
3.7356 | 650 | 0.0001 | - |
4.0230 | 700 | 0.0001 | - |
4.3103 | 750 | 0.0001 | - |
4.5977 | 800 | 0.0001 | - |
4.8851 | 850 | 0.0001 | - |
5.1724 | 900 | 0.0 | - |
5.4598 | 950 | 0.0 | - |
5.7471 | 1000 | 0.0 | - |
6.0345 | 1050 | 0.0 | - |
6.3218 | 1100 | 0.0 | - |
6.6092 | 1150 | 0.0 | - |
6.8966 | 1200 | 0.0 | - |
7.1839 | 1250 | 0.0 | - |
7.4713 | 1300 | 0.0001 | - |
7.7586 | 1350 | 0.0 | - |
8.0460 | 1400 | 0.0 | - |
8.3333 | 1450 | 0.0 | - |
8.6207 | 1500 | 0.0 | - |
8.9080 | 1550 | 0.0 | - |
9.1954 | 1600 | 0.0 | - |
9.4828 | 1650 | 0.0 | - |
9.7701 | 1700 | 0.0 | - |
10.0575 | 1750 | 0.0 | - |
10.3448 | 1800 | 0.0 | - |
10.6322 | 1850 | 0.0 | - |
10.9195 | 1900 | 0.0 | - |
11.2069 | 1950 | 0.0 | - |
11.4943 | 2000 | 0.0 | - |
11.7816 | 2050 | 0.0 | - |
12.0690 | 2100 | 0.0 | - |
12.3563 | 2150 | 0.0 | - |
12.6437 | 2200 | 0.0 | - |
12.9310 | 2250 | 0.0 | - |
13.2184 | 2300 | 0.0 | - |
13.5057 | 2350 | 0.0 | - |
13.7931 | 2400 | 0.0 | - |
14.0805 | 2450 | 0.0 | - |
14.3678 | 2500 | 0.0 | - |
14.6552 | 2550 | 0.0 | - |
14.9425 | 2600 | 0.0 | - |
15.2299 | 2650 | 0.0 | - |
15.5172 | 2700 | 0.0 | - |
15.8046 | 2750 | 0.0 | - |
16.0920 | 2800 | 0.0 | - |
16.3793 | 2850 | 0.0 | - |
16.6667 | 2900 | 0.0 | - |
16.9540 | 2950 | 0.0 | - |
17.2414 | 3000 | 0.0 | - |
17.5287 | 3050 | 0.0 | - |
17.8161 | 3100 | 0.0 | - |
18.1034 | 3150 | 0.0 | - |
18.3908 | 3200 | 0.0 | - |
18.6782 | 3250 | 0.0 | - |
18.9655 | 3300 | 0.0 | - |
19.2529 | 3350 | 0.0 | - |
19.5402 | 3400 | 0.0 | - |
19.8276 | 3450 | 0.0 | - |
20.1149 | 3500 | 0.0 | - |
20.4023 | 3550 | 0.0 | - |
20.6897 | 3600 | 0.0 | - |
20.9770 | 3650 | 0.0 | - |
21.2644 | 3700 | 0.0 | - |
21.5517 | 3750 | 0.0 | - |
21.8391 | 3800 | 0.0 | - |
22.1264 | 3850 | 0.0 | - |
22.4138 | 3900 | 0.0 | - |
22.7011 | 3950 | 0.0 | - |
22.9885 | 4000 | 0.0 | - |
23.2759 | 4050 | 0.0 | - |
23.5632 | 4100 | 0.0 | - |
23.8506 | 4150 | 0.0 | - |
24.1379 | 4200 | 0.0 | - |
24.4253 | 4250 | 0.0 | - |
24.7126 | 4300 | 0.0 | - |
25.0 | 4350 | 0.0 | - |
25.2874 | 4400 | 0.0 | - |
25.5747 | 4450 | 0.0 | - |
25.8621 | 4500 | 0.0 | - |
26.1494 | 4550 | 0.0 | - |
26.4368 | 4600 | 0.0 | - |
26.7241 | 4650 | 0.0 | - |
27.0115 | 4700 | 0.0 | - |
27.2989 | 4750 | 0.0 | - |
27.5862 | 4800 | 0.0 | - |
27.8736 | 4850 | 0.0 | - |
28.1609 | 4900 | 0.0 | - |
28.4483 | 4950 | 0.0 | - |
28.7356 | 5000 | 0.0 | - |
29.0230 | 5050 | 0.0 | - |
29.3103 | 5100 | 0.0 | - |
29.5977 | 5150 | 0.0 | - |
29.8851 | 5200 | 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}
}