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: 5 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	'(신세계김해점)에트로 프로푸미 헤어밴드 01046 05 1099 ONE SIZE 신세계백화점' 'Baby scrunchie 3set (White/Beige/Black) 빌라드실크 곱창밴드 미니 실크 스크런치 세트 주식회사 실크랩' '간단 헤어밴드 미키마우스 머리띠 왕 리본 남자 캐릭터 플라스틱 반짝이 1-4. 글리터 / 블랙 아이드림'
1.0	'위즈템 헤어밴드 진주 크리스탈 머리끈 연핑크 파파닐' '둥근고무줄 (대용량) 칼라 금 은 천고무줄 벌크 탄성끈 가는줄 /굵은줄 02. 대용량 굵은줄(2.5mmx60M)_금색 마이1004(MY1004)' '천연 컬러 고무 끈 고무줄 생활용품 3M 하늘색 제이앤제이웍스'
0.0	'인모 남자가발 정수리 커버 자연스러운 O형 커버가발 마오_인모14X14 하이윤' '얼굴소멸 히메컷 가발 앞머리 사이드뱅 옆머리 부분 히메컷 사이드뱅 옆2p-내츄럴브라운 와우마켓' '얼굴소멸 히메컷 가발 앞머리 사이드뱅 옆머리 부분 옆2p-라이트브라운 이지구'
4.0	'무지 12컬러 심플 리본 바나나핀 핫핑크 하얀당나귀' '네임핀/이름핀/네임브로치/어린이집선물/유치원선물 5글자(영어6자~8자)_별_브로치 쭈스타' '메탈 셀룰로오스 꼬임 올림머리 집게핀 사각4170_아이스옐로우 엑스엔서'
3.0	'웨딩 드레스 유니크 베일 셀프 촬영 소품 대형 리본 잡지 모델 패션쇼 장식 액세서리 머리 04.파란 (핸드메이드) 더비공이(TheB02)' '슈퍼 요정 흰색 보석 웨딩 헤어 타워 공연 여행 T15-a_선택하세요 아토버디' '뿌리볼륨집게3p 건강드림'

Evaluation

Metrics

Label	Metric
all	0.9541

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_ac16")
# Run inference
preds = model("파티 벨벳 심플 왕리본핀 반묶음핀 30칼라 와인_납작핀대 릴리트리")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	9.956	24

Label	Training Sample Count
0.0	50
1.0	50
2.0	50
3.0	50
4.0	50

Training Hyperparameters

batch_size: (512, 512)
num_epochs: (20, 20)
max_steps: -1
sampling_strategy: oversampling
num_iterations: 40
body_learning_rate: (2e-05, 2e-05)
head_learning_rate: 2e-05
loss: CosineSimilarityLoss
distance_metric: cosine_distance
margin: 0.25
end_to_end: False
use_amp: False
warmup_proportion: 0.1
seed: 42
eval_max_steps: -1
load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.025	1	0.4499	-
1.25	50	0.2065	-
2.5	100	0.0446	-
3.75	150	0.0001	-
5.0	200	0.0	-
6.25	250	0.0001	-
7.5	300	0.0	-
8.75	350	0.0	-
10.0	400	0.0	-
11.25	450	0.0	-
12.5	500	0.0	-
13.75	550	0.0	-
15.0	600	0.0	-
16.25	650	0.0	-
17.5	700	0.0	-
18.75	750	0.0	-
20.0	800	0.0	-

Framework Versions

Python: 3.10.12
SetFit: 1.1.0.dev0
Sentence Transformers: 3.1.1
Transformers: 4.46.1
PyTorch: 2.4.0+cu121
Datasets: 2.20.0
Tokenizers: 0.20.0

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}
}

mini1013
/

master_cate_ac16