intvy/setfit-absa-all-MiniLM-L6-v2-laptops-aspect

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: joshuasundance/setfit-absa-all-MiniLM-L6-v2-laptops-aspect
• SetFitABSA Polarity Model: joshuasundance/setfit-absa-all-mpnet-base-v2-laptops-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	'cord:I charge it at night and skip taking the cord with me because of the good battery life.' 'battery life:I charge it at night and skip taking the cord with me because of the good battery life.' 'service center:The tech guy then said the service center does not do 1-to-1 exchange and I have to direct my concern to the "sales" team, which is the retail shop which I bought my netbook from.'
no aspect	'night:I charge it at night and skip taking the cord with me because of the good battery life.' 'skip:I charge it at night and skip taking the cord with me because of the good battery life.' 'exchange:The tech guy then said the service center does not do 1-to-1 exchange and I have to direct my concern to the "sales" team, which is the retail shop which I bought my netbook from.'

Evaluation

Metrics

Label	Accuracy
all	0.8240

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(
    "joshuasundance/setfit-absa-all-MiniLM-L6-v2-laptops-aspect",
    "joshuasundance/setfit-absa-all-mpnet-base-v2-laptops-polarity",
    spacy_model="en_core_web_sm",
)
# Run inference
preds = model("This laptop meets every expectation and Windows 7 is great!")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	2	21.1510	42

Label	Training Sample Count
no aspect	119
aspect	126

Training Hyperparameters

• batch_size: (128, 128)
• 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
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0042	1	0.3776	-
0.2110	50	0.2644	0.2622
0.4219	100	0.2248	0.2437
0.6329	150	0.0059	0.2238
0.8439	200	0.0017	0.2326
1.0549	250	0.0012	0.2382
1.2658	300	0.0008	0.2455
1.4768	350	0.0006	0.2328
1.6878	400	0.0005	0.243

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.7
• SetFit: 1.0.3
• Sentence Transformers: 2.3.0
• spaCy: 3.7.2
• Transformers: 4.37.2
• PyTorch: 2.1.2+cu118
• Datasets: 2.16.1
• Tokenizers: 0.15.1

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}
}