BLigtenberg/setfit-mpnet-final

SetFit with sentence-transformers/all-mpnet-base-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-mpnet-base-v2 as the Sentence Transformer embedding model. A OneVsRestClassifier 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-mpnet-base-v2
• Classification head: a OneVsRestClassifier instance
• Maximum Sequence Length: 384 tokens

Model Sources

• Repository: SetFit on GitHub
• Paper: Efficient Few-Shot Learning Without Prompts
• Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

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("I'm so scared help me please [SEP] I CAN'T STOP THINKING ABOUT DEATH AND GETTING OLD! I'm 16 and i keep thinking that this is as good as my life is going to get(not that its been all that great) I can't,stop thinking about hos inevitably we all fucking die. When i watch a movie i think about how old it is. When i listen to old bands i think about how they've slowly,declined and faded into obscurity. I want to enjoy the present but I cant stop thinking about getting old and dying.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	52	173.5532	416

Training Hyperparameters

• batch_size: (16, 16)
• num_epochs: (5, 5)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 20
• 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
0.0043	1	0.183	-
0.2128	50	0.191	-
0.4255	100	0.1144	-
0.6383	150	0.0475	-
0.8511	200	0.0341	-
1.0638	250	0.0204	-
1.2766	300	0.0191	-
1.4894	350	0.019	-
1.7021	400	0.0144	-
1.9149	450	0.014	-
2.1277	500	0.0118	-
2.3404	550	0.0145	-
2.5532	600	0.0116	-
2.7660	650	0.0119	-
2.9787	700	0.0137	-
3.1915	750	0.0105	-
3.4043	800	0.0108	-
3.6170	850	0.0124	-
3.8298	900	0.0128	-
4.0426	950	0.0117	-
4.2553	1000	0.0104	-
4.4681	1050	0.0101	-
4.6809	1100	0.01	-
4.8936	1150	0.0097	-

Framework Versions

• Python: 3.10.12
• SetFit: 1.1.3
• Sentence Transformers: 3.3.1
• Transformers: 4.46.3
• PyTorch: 2.7.1+cu118
• Datasets: 5.0.0
• Tokenizers: 0.20.3

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}
}