Sfekih/sentence_relevancy_model

SetFit with sentence-transformers/all-MiniLM-L6-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-MiniLM-L6-v2 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: sentence-transformers/all-MiniLM-L6-v2
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 256 tokens
• Number of Classes: 2 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
1	'In addition to date, UNFPA has distributed dignity kits to 12,650 people through partners.' 'In particular, WHO, acting on the eight pillars of the global WHO Strategic Preparedness and Response Plan, continues engaging the MoH and health partners to enhance technical capacity and awareness, including on rational use of PPEs, case management, infection prevention and control, environmental disinfection, and risk communication; and is focused on procuring and enhancing integral medical supplies including in laboratory testing and PPE for case management and healthcare facilities' 'Adicionalmente, la propuesta incluyóla entrega de mercados para asistencia alimentaria al menos a 244 personas sobrevivientes de Minas Antipersonal (MAP), Municiones sin Explotar (MSE) y/o Artefactos Explosivos Improvisados (AEI) y sus núcleos familiares.'
0	'Labor market indicators by age 42 List of figures Figure 2.' 'Women’s involvement in conflict mediation: percentage of women leading initiatives 52 List of boxes Box 2.' 'Entrevista telefónica, funcionario de la ONU, octubre de 2023.'

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("setfit_model_id")
# Run inference
preds = model("Consulte los materiales adjuntos para lecturas adicionales.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	2	24.6961	85

Label	Training Sample Count
0	81
1	100

Training Hyperparameters

• batch_size: (32, 32)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 35
• 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
• l2_weight: 0.01
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0025	1	0.3104	-
0.1263	50	0.2567	-
0.2525	100	0.0406	-
0.3788	150	0.0034	-
0.5051	200	0.0017	-
0.6313	250	0.0012	-
0.7576	300	0.0009	-
0.8838	350	0.0008	-

Framework Versions

• Python: 3.11.5
• SetFit: 1.1.0
• Sentence Transformers: 3.1.1
• Transformers: 4.45.1
• PyTorch: 2.1.0
• Datasets: 2.17.1
• 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}
}