metadata
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 아레나 모던 대리석 다용도 도어 수납장 가구/인테리어>수납가구>수납장
- text: 원목 문갑 약장 자개농 엔틱 고가구 거실 인테리어 선반 가구/인테리어>수납가구>고가구
- text: 아카시아 대용량 2단 선반 행거 400호 가구/인테리어>수납가구>행거
- text: 수납 박스 우드 원목 상자 케이스 나무함 정리함 바늘 실 가구/인테리어>수납가구>소품수납함
- text: 엽서 진열대 전단지 전시대 디스플레이 회전 가판대 매거진 가구/인테리어>수납가구>잡지꽂이
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: 12 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 |
---|---|
10.0 |
|
0.0 |
|
3.0 |
|
11.0 |
|
4.0 |
|
2.0 |
|
7.0 |
|
8.0 |
|
6.0 |
|
5.0 |
|
1.0 |
|
9.0 |
|
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_fi5")
# Run inference
preds = model("아카시아 대용량 2단 선반 행거 400호 가구/인테리어>수납가구>행거")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 9.0095 | 20 |
Label | Training Sample Count |
---|---|
0.0 | 69 |
1.0 | 70 |
2.0 | 70 |
3.0 | 70 |
4.0 | 70 |
5.0 | 70 |
6.0 | 70 |
7.0 | 70 |
8.0 | 70 |
9.0 | 69 |
10.0 | 70 |
11.0 | 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 |
---|---|---|---|
0.0061 | 1 | 0.4852 | - |
0.3049 | 50 | 0.4994 | - |
0.6098 | 100 | 0.4134 | - |
0.9146 | 150 | 0.1731 | - |
1.2195 | 200 | 0.0287 | - |
1.5244 | 250 | 0.0058 | - |
1.8293 | 300 | 0.0003 | - |
2.1341 | 350 | 0.0001 | - |
2.4390 | 400 | 0.0001 | - |
2.7439 | 450 | 0.0001 | - |
3.0488 | 500 | 0.0 | - |
3.3537 | 550 | 0.0 | - |
3.6585 | 600 | 0.0 | - |
3.9634 | 650 | 0.0 | - |
4.2683 | 700 | 0.0 | - |
4.5732 | 750 | 0.0 | - |
4.8780 | 800 | 0.0 | - |
5.1829 | 850 | 0.0 | - |
5.4878 | 900 | 0.0 | - |
5.7927 | 950 | 0.0 | - |
6.0976 | 1000 | 0.0 | - |
6.4024 | 1050 | 0.0 | - |
6.7073 | 1100 | 0.0 | - |
7.0122 | 1150 | 0.0002 | - |
7.3171 | 1200 | 0.0001 | - |
7.6220 | 1250 | 0.0 | - |
7.9268 | 1300 | 0.0 | - |
8.2317 | 1350 | 0.0 | - |
8.5366 | 1400 | 0.0 | - |
8.8415 | 1450 | 0.0 | - |
9.1463 | 1500 | 0.0 | - |
9.4512 | 1550 | 0.0 | - |
9.7561 | 1600 | 0.0 | - |
10.0610 | 1650 | 0.0 | - |
10.3659 | 1700 | 0.0 | - |
10.6707 | 1750 | 0.0 | - |
10.9756 | 1800 | 0.0 | - |
11.2805 | 1850 | 0.0 | - |
11.5854 | 1900 | 0.0 | - |
11.8902 | 1950 | 0.0 | - |
12.1951 | 2000 | 0.0 | - |
12.5 | 2050 | 0.0 | - |
12.8049 | 2100 | 0.0 | - |
13.1098 | 2150 | 0.0 | - |
13.4146 | 2200 | 0.0 | - |
13.7195 | 2250 | 0.0 | - |
14.0244 | 2300 | 0.0 | - |
14.3293 | 2350 | 0.0 | - |
14.6341 | 2400 | 0.0 | - |
14.9390 | 2450 | 0.0 | - |
15.2439 | 2500 | 0.0 | - |
15.5488 | 2550 | 0.0 | - |
15.8537 | 2600 | 0.0 | - |
16.1585 | 2650 | 0.0 | - |
16.4634 | 2700 | 0.0 | - |
16.7683 | 2750 | 0.0 | - |
17.0732 | 2800 | 0.0 | - |
17.3780 | 2850 | 0.0 | - |
17.6829 | 2900 | 0.0 | - |
17.9878 | 2950 | 0.0 | - |
18.2927 | 3000 | 0.0 | - |
18.5976 | 3050 | 0.0 | - |
18.9024 | 3100 | 0.0 | - |
19.2073 | 3150 | 0.0 | - |
19.5122 | 3200 | 0.0 | - |
19.8171 | 3250 | 0.0 | - |
20.1220 | 3300 | 0.0 | - |
20.4268 | 3350 | 0.0 | - |
20.7317 | 3400 | 0.0 | - |
21.0366 | 3450 | 0.0 | - |
21.3415 | 3500 | 0.0 | - |
21.6463 | 3550 | 0.0 | - |
21.9512 | 3600 | 0.0 | - |
22.2561 | 3650 | 0.0 | - |
22.5610 | 3700 | 0.0 | - |
22.8659 | 3750 | 0.0 | - |
23.1707 | 3800 | 0.0 | - |
23.4756 | 3850 | 0.0 | - |
23.7805 | 3900 | 0.0 | - |
24.0854 | 3950 | 0.0 | - |
24.3902 | 4000 | 0.0 | - |
24.6951 | 4050 | 0.0 | - |
25.0 | 4100 | 0.0 | - |
25.3049 | 4150 | 0.0 | - |
25.6098 | 4200 | 0.0 | - |
25.9146 | 4250 | 0.0 | - |
26.2195 | 4300 | 0.0 | - |
26.5244 | 4350 | 0.0 | - |
26.8293 | 4400 | 0.0 | - |
27.1341 | 4450 | 0.0 | - |
27.4390 | 4500 | 0.0 | - |
27.7439 | 4550 | 0.0 | - |
28.0488 | 4600 | 0.0 | - |
28.3537 | 4650 | 0.0 | - |
28.6585 | 4700 | 0.0 | - |
28.9634 | 4750 | 0.0 | - |
29.2683 | 4800 | 0.0 | - |
29.5732 | 4850 | 0.0 | - |
29.8780 | 4900 | 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}
}