mitra-mir/setfit-model-ESG-governance

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	'Our goal is for every employee to feel a strong sense of belonging and psychological safety.' 'Revenue from products labeled andor marketed to promote health and nutrition attributes is approximately 10.3 billion.' 'Assess climate change scenarios of key material risks.'
1.0	'We assessed accounting estimates for bias and evaluated whether the circumstances producing the bias, if any, represent a risk of material misstatement due to fraud.' 'Board to review and make recommendations to shareholders concerning the composition of the Board of Directors, with particular focus on issues of independence.' 'Our group audit mainly focused on significant group entities in terms of size and financial interest or where significant risks or complex activities were present, leading to full scope audits performed for including 2 nonconsolidated components.'

Evaluation

Metrics

Label	Accuracy
all	0.8262

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-governance")
# Run inference
preds = model("The Company will also make every effort to improve the effectiveness of its sustainability reporting.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	2	24.51	74

Label	Training Sample Count
0.0	146
1.0	54

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
0.002	1	0.479	-
0.1	50	0.2577	-
0.2	100	0.0589	-
0.3	150	0.0008	-
0.4	200	0.0004	-
0.5	250	0.0003	-
0.6	300	0.0002	-
0.7	350	0.0002	-
0.8	400	0.0002	-
0.9	450	0.0002	-
1.0	500	0.0002	-

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}
}