Kevinger/setfit-hub-multilabel-example

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

This is a SetFit model trained on the Kevinger/hub-report-dataset dataset that can be used for Text Classification. This SetFit model uses sentence-transformers/paraphrase-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/paraphrase-mpnet-base-v2
• Classification head: a OneVsRestClassifier instance
• Maximum Sequence Length: 512 tokens
• Training Dataset: Kevinger/hub-report-dataset

Model Sources

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

Evaluation

Metrics

Label	Accuracy
all	0.6529

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("Kevinger/setfit-hub-multilabel-example")
# Run inference
preds = model("LEVERETT — Dakin Humane Society announced Wednesday that it has sold its former animal shelter at 63 Montague Road in Leverett to Better Together Dog Rescue.

The news release didn’t include a sales price for the 3,480-square-foot building on 5 acres of land.

But records at the Franklin County Registry of Deeds show the sale was for $575,000.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	53	386.3906	2161

Training Hyperparameters

• batch_size: (8, 8)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 75
• 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
0.0008	1	0.1304	-
0.0417	50	0.1596	-
0.0833	100	0.132	-
0.125	150	0.0064	-
0.1667	200	0.0017	-
0.2083	250	0.0004	-
0.25	300	0.0001	-
0.2917	350	0.0002	-
0.3333	400	0.0003	-
0.375	450	0.0002	-
0.4167	500	0.0001	-
0.4583	550	0.0002	-
0.5	600	0.0002	-
0.5417	650	0.0002	-
0.5833	700	0.0001	-
0.625	750	0.0001	-
0.6667	800	0.0001	-
0.7083	850	0.0001	-
0.75	900	0.0	-
0.7917	950	0.0001	-
0.8333	1000	0.0001	-
0.875	1050	0.0001	-
0.9167	1100	0.0001	-
0.9583	1150	0.0	-
1.0	1200	0.0001	-

Framework Versions

• Python: 3.10.12
• SetFit: 1.0.3
• Sentence Transformers: 2.3.1
• Transformers: 4.35.2
• PyTorch: 2.1.0+cu121
• Datasets: 2.17.0
• Tokenizers: 0.15.2

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}
}