omymble/setfit-absa-books-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_lg
• SetFitABSA Aspect Model: omymble/setfit-absa-books-aspect
• SetFitABSA Polarity Model: omymble/setfit-absa-books-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	'Poirot:After reading nearly 30 novels, Poirot had been a part of life' 'Michael Dobbs:The cast of characters in this sweeping story by Michael Dobbs of political maneuvering, skullduggery, and backstabbing is an historical Who's Who of the times: the ailing, haughty, and pacifist Chamberlain, who personifies England's bitter memories of the Great War and the popular concept of "never again"; the ambitious and self-absorbed Churchill, whose pugnacity sometimes clouds prudence; the defeatist, philandering, and anti-Semitic U' "Jack:Jack is a wonderful beleaguered hero who starts off by quickly realizing he don't know jack even about himself and as he investigates realizes each new clue proves he knows even less than he thought"
no aspect	'novels:After reading nearly 30 novels, Poirot had been a part of life' 'part:After reading nearly 30 novels, Poirot had been a part of life' 'life:After reading nearly 30 novels, Poirot had been a part of life'

Evaluation

Metrics

Label	Accuracy
all	0.8542

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(
    "omymble/setfit-absa-books-aspect",
    "omymble/setfit-absa-books-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	6	34.7122	79

Label	Training Sample Count
no aspect	280
aspect	57

Training Hyperparameters

• batch_size: (256, 256)
• num_epochs: (2, 2)
• 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
0.0031	1	0.3698	-
0.1558	50	0.3449	0.3303
0.3115	100	0.3032	0.294
0.4673	150	0.2878	0.266
0.6231	200	0.2414	0.2535
0.7788	250	0.2456	0.2494
0.9346	300	0.2374	0.2477
1.0903	350	0.2407	0.2472
1.2461	400	0.2406	0.2467
1.4019	450	0.2276	0.2465
1.5576	500	0.2248	0.2465
1.7134	550	0.2241	0.2464
1.8692	600	0.2245	0.2463

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• SetFit: 1.0.3
• Sentence Transformers: 3.0.1
• spaCy: 3.7.4
• Transformers: 4.39.0
• PyTorch: 2.3.1+cu121
• Datasets: 2.20.0
• Tokenizers: 0.15.2

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}
}