mini1013/master_cate_sl29

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:

1. Fine-tuning a Sentence Transformer with contrastive learning.
2. 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: 7 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
2.0	'제맥스 박서 야외용 탁구대 스포츠/레저>탁구>탁구대' '챔피온 H-50 탁구대 스포츠/레저>탁구>탁구대' '휠러스 미니 탁구대 스포츠/레저>탁구>탁구대'
3.0	'버터플라이 RDJ S1 탁구라켓 스포츠/레저>탁구>탁구라켓' '엑시옴 하야부사 ZL PRO 탁구라켓 FL 스포츠/레저>탁구>탁구라켓' '고집통 나노100 탁구라켓 스포츠/레저>탁구>탁구라켓'
5.0	'미즈노 탁구복 게임 팬츠 탁구 유니섹스 Mizuno 82JBA10009 스포츠/레저>탁구>탁구의류' '버터플라이 윈로고 티셔츠 스포츠/레저>탁구>탁구의류' '미즈노 탁구복 게임 바지 탁구웨어 426388 82JB900109 스포츠/레저>탁구>탁구의류'
6.0	'리닝 탁구화 23 아틀란티스 보아 라임 - 바운스플러스 슈퍼라이트 시리즈 스포츠/레저>탁구>탁구화' '아식스 탁구화 ATTACK DOMINATE FF 2 1073A010-003 스포츠/레저>탁구>탁구화' '미즈노 탁구화 웨이브 드라이브 EL 경량성 쿠션성 81GA2001 스포츠/레저>탁구>탁구화'
0.0	'엑시옴 솔라이트 SOLITE 백팩 스포츠/레저>탁구>기타탁구용품' 'Dawei 다웨이 미디엄 핌플러버 스폰지버전 - 탁구러버 388C-1 스포츠/레저>탁구>기타탁구용품' '도닉 탁구 러버 블루스톰 프로 스포츠/레저>탁구>기타탁구용품'
1.0	'비코트 슈퍼 ABS 40+ 탁구공 스포츠/레저>탁구>탁구공' '안드로 스피드볼 Mi 1 탁구공 스포츠/레저>탁구>탁구공' '오로라 AURORA 3성 BST 시합구 스포츠/레저>탁구>탁구공'
4.0	'탁구러버 티바 에볼루션 MX-P 스포츠/레저>탁구>탁구러버' '티바 하이브리드 K3 HYBRID K3 평면 탁구러버 스포츠/레저>탁구>탁구러버' '엑시옴 탁구러버 베가 프로 스포츠/레저>탁구>탁구러버'

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_sl29")
# Run inference
preds = model("엑시옴 탁구상의 토마스 탁구유니폼 티셔츠 스포츠/레저>탁구>탁구의류")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	6.8776	14

Label	Training Sample Count
0.0	70
1.0	70
2.0	25
3.0	70
4.0	9
5.0	70
6.0	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
0.0133	1	0.4688	-
0.6667	50	0.5004	-
1.3333	100	0.1817	-
2.0	150	0.0186	-
2.6667	200	0.0024	-
3.3333	250	0.0009	-
4.0	300	0.0001	-
4.6667	350	0.0	-
5.3333	400	0.0	-
6.0	450	0.0	-
6.6667	500	0.0	-
7.3333	550	0.0	-
8.0	600	0.0	-
8.6667	650	0.0	-
9.3333	700	0.0	-
10.0	750	0.0	-
10.6667	800	0.0	-
11.3333	850	0.0	-
12.0	900	0.0	-
12.6667	950	0.0	-
13.3333	1000	0.0	-
14.0	1050	0.0	-
14.6667	1100	0.0	-
15.3333	1150	0.0	-
16.0	1200	0.0	-
16.6667	1250	0.0	-
17.3333	1300	0.0	-
18.0	1350	0.0	-
18.6667	1400	0.0	-
19.3333	1450	0.0	-
20.0	1500	0.0	-
20.6667	1550	0.0	-
21.3333	1600	0.0	-
22.0	1650	0.0	-
22.6667	1700	0.0	-
23.3333	1750	0.0	-
24.0	1800	0.0	-
24.6667	1850	0.0	-
25.3333	1900	0.0	-
26.0	1950	0.0	-
26.6667	2000	0.0	-
27.3333	2050	0.0	-
28.0	2100	0.0	-
28.6667	2150	0.0	-
29.3333	2200	0.0	-
30.0	2250	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}
}