najwaa/absa-laptops-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-laptops-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	'light weight:this computer is so light weight and easy to carry.' 'carry:this computer is so light weight and easy to carry.' 'lightweight:very lightweight.'
no aspect	'computer:this computer is so light weight and easy to carry.' "premium:The build quality feels premium but it's surprisingly heavy for daily commuting." "commuting:The build quality feels premium but it's surprisingly heavy for daily commuting."

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-laptops-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	2	18.3046	52

Label	Training Sample Count
no aspect	341
aspect	263

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
0.0002	1	0.2288	-
0.0017	10	-	0.3073
0.0034	20	-	0.3065
0.0052	30	-	0.3052
0.0069	40	-	0.3034
0.0086	50	0.3393	0.3011
0.0103	60	-	0.2984
0.0120	70	-	0.2954
0.0138	80	-	0.2923
0.0155	90	-	0.2888
0.0172	100	0.3016	0.2855
0.0189	110	-	0.2821
0.0206	120	-	0.2786
0.0224	130	-	0.2754
0.0241	140	-	0.2722
0.0258	150	0.2875	0.2684
0.0275	160	-	0.2653
0.0292	170	-	0.2629
0.0310	180	-	0.2611
0.0327	190	-	0.2597
0.0344	200	0.272	0.2588
0.0361	210	-	0.2585
0.0378	220	-	0.2584
0.0396	230	-	0.2584
0.0413	240	-	0.2584
0.0430	250	0.2652	0.2585
0.0447	260	-	0.2581
0.0464	270	-	0.2580
0.0482	280	-	0.2580
0.0499	290	-	0.2580
0.0516	300	0.2635	0.2578
0.0533	310	-	0.2578
0.0550	320	-	0.2575
0.0567	330	-	0.2572
0.0585	340	-	0.2573
0.0602	350	0.2608	0.2572
0.0619	360	-	0.2576
0.0636	370	-	0.2576
0.0653	380	-	0.2579

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}
}