najwaa/absa-headphones-aspect-p2

SetFit Aspect Model with sentence-transformers/all-MiniLM-L6-v2

This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses sentence-transformers/all-MiniLM-L6-v2 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification. In particular, this model is in charge of filtering aspect span candidates.

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.

This model was trained within the context of a larger system for ABSA, which looks like so:

1. Use a spaCy model to select possible aspect span candidates.
2. Use this SetFit model to filter these possible aspect span candidates.
3. Use a SetFit model to classify the filtered aspect span candidates.

Model Details

Model Description

• Model Type: SetFit
• Sentence Transformer body: sentence-transformers/all-MiniLM-L6-v2
• Classification head: a LogisticRegression instance
• spaCy Model: en_core_web_sm
• SetFitABSA Aspect Model: najwaa/absa-headphones-aspect-p2
• SetFitABSA Polarity Model: setfit-absa-polarity
• 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

Model Labels

Label	Examples
aspect	'sound quality:Amazing sound quality with deep bass that really makes music come alive.' 'bass:Amazing sound quality with deep bass that really makes music come alive.' 'audio:The audio is crystal clear but they become uncomfortable after wearing for more than an hour.'
no aspect	'music:Amazing sound quality with deep bass that really makes music come alive.' 'crystal:The audio is crystal clear but they become uncomfortable after wearing for more than an hour.' 'hour:The audio is crystal clear but they become uncomfortable after wearing for more than an hour.'

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 AbsaModel

# Download from the 🤗 Hub
model = AbsaModel.from_pretrained(
    "najwaa/absa-headphones-aspect-p2",
    "setfit-absa-polarity",
)
# Run inference
preds = model("The food was great, but the venue is just way too busy.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	4	18.3499	52

Label	Training Sample Count
no aspect	271
aspect	152

Training Hyperparameters

• batch_size: (32, 32)
• num_epochs: (5, 5)
• max_steps: -1
• sampling_strategy: oversampling
• 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: True
• warmup_proportion: 0.1
• l2_weight: 0.01
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0003	1	0.3624	-
0.0033	10	-	0.3137
0.0066	20	-	0.3121
0.0099	30	-	0.3096
0.0132	40	-	0.3062
0.0165	50	0.3537	0.3019
0.0198	60	-	0.2968
0.0231	70	-	0.2916
0.0264	80	-	0.2862
0.0297	90	-	0.2808
0.0330	100	0.309	0.2757
0.0363	110	-	0.2711
0.0396	120	-	0.2671
0.0429	130	-	0.2640
0.0462	140	-	0.2625
0.0495	150	0.2754	0.2617
0.0528	160	-	0.2618
0.0561	170	-	0.2618
0.0594	180	-	0.2619
0.0627	190	-	0.2624
0.0660	200	0.2608	0.2618

Framework Versions

• Python: 3.11.12
• SetFit: 1.1.2
• Sentence Transformers: 4.1.0
• spaCy: 3.7.5
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu124
• Datasets: 3.6.0
• Tokenizers: 0.21.1

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}
}