Zlovoblachko/dimension1_setfit

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

Evaluation

Metrics

Label	F1
all	0.7727

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("Zlovoblachko/dimension1_setfit")
# Run inference
preds = model("I loved the spiderman movie!")

Training Details

Training Hyperparameters

• batch_size: (16, 16)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: oversampling
• body_learning_rate: (4.4226261631087265e-05, 4.4226261631087265e-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
0.0006	1	0.2748	-
0.0280	50	0.2678	-
0.0559	100	0.2688	-
0.0839	150	0.2709	-
0.1119	200	0.2656	-
0.1398	250	0.259	-
0.1678	300	0.2565	-
0.1957	350	0.2655	-
0.2237	400	0.2737	-
0.2517	450	0.2501	-
0.2796	500	0.2512	-
0.3076	550	0.2381	-
0.3356	600	0.2568	-
0.3635	650	0.2642	-
0.3915	700	0.2743	-
0.4195	750	0.2635	-
0.4474	800	0.263	-
0.4754	850	0.2541	-
0.5034	900	0.2492	-
0.5313	950	0.26	-
0.5593	1000	0.257	-
0.5872	1050	0.2525	-
0.6152	1100	0.2594	-
0.6432	1150	0.2656	-
0.6711	1200	0.2737	-
0.6991	1250	0.2683	-
0.7271	1300	0.259	-
0.7550	1350	0.2617	-
0.7830	1400	0.294	-
0.8110	1450	0.2446	-
0.8389	1500	0.2618	-
0.8669	1550	0.2562	-
0.8949	1600	0.264	-
0.9228	1650	0.2534	-
0.9508	1700	0.2484	-
0.9787	1750	0.2666	-

Framework Versions

• Python: 3.10.12
• SetFit: 1.1.0
• Sentence Transformers: 3.2.1
• Transformers: 4.44.2
• PyTorch: 2.5.0+cu121
• Datasets: 3.0.2
• 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}
}