mitra-mir/setfit-model-ESG-social

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

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/paraphrase-mpnet-base-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/paraphrase-mpnet-base-v2
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 512 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
0.0	'The concentration of our sales in our fourth fiscal quarter increases this impact as the revenue impact of most fourth fiscal quarter subscription sales will not be realized until the following fiscal year.' 'The deposit insurance fund of the FDIC insures deposit accounts in HomeTrust Bank up to 250,000 per separately insured deposit ownership right or category.' 'On 19 October 2020 the Company entered into a 25 million revolving credit facility agreement with State Street Bank International GmbH.'
1.0	'The IH Laboratory supports our field organization and offers our customers a wide array of sampling and analysis protocols based on Occupational Safety and Health Administration (OSHA) and National Institute for Occupational Safety and Health methodologies.' 'We partnered with the Fair Wage Network to increase our understanding of the different definitions of fair wages and to benchmark against their extensive fair and living wage database.' 'Strategy is to ensure all our employees worldwide earn at least a fair wage.'

Evaluation

Metrics

Label	Accuracy
all	0.7909

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("mitra-mir/setfit-model-ESG-social")
# Run inference
preds = model("and importance of the climate crisis requires everyone to play their part.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	4	24.5578	112

Label	Training Sample Count
0.0	147
1.0	52

Training Hyperparameters

• batch_size: (16, 16)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 20
• 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
0.0020	1	0.4072	-
0.1004	50	0.2038	-
0.2008	100	0.0155	-
0.3012	150	0.0003	-
0.4016	200	0.0002	-
0.5020	250	0.0002	-
0.6024	300	0.0002	-
0.7028	350	0.0001	-
0.8032	400	0.0001	-
0.9036	450	0.0001	-

Framework Versions

• Python: 3.11.6
• SetFit: 1.1.0
• Sentence Transformers: 3.2.1
• Transformers: 4.43.4
• PyTorch: 2.4.1+cu121
• Datasets: 3.0.1
• 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}
}