SetFit Polarity Model with BAAI/bge-small-en-v1.5

This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses BAAI/bge-small-en-v1.5 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification. In particular, this model is in charge of classifying aspect polarities.

The model has been trained using an efficient few-shot learning technique that involves:

Fine-tuning a Sentence Transformer with contrastive learning.
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:

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

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: BAAI/bge-small-en-v1.5
Classification head: a LogisticRegression instance
spaCy Model: en_core_web_lg
SetFitABSA Aspect Model: omymble/books-full-bge-aspect
SetFitABSA Polarity Model: omymble/books-full-bge-polarity
Maximum Sequence Length: 512 tokens
Number of Classes: 3 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
negative	"too dark for younger ones, unless you:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!" 'The mystery is secondary to:The mystery is secondary to the rest of the story and is only really approached in the remaining 30 pages of the book' 'was only my book with this problem:I have no idea if it was only my book with this problem'
neutral	'world, as Nix weaves a wonderful:-enjoy the genre of fantasies, of a unknown world, as Nix weaves a wonderful tale of the things that will open your eyes to a different world' 'Arthur must get through:Arthur must get through some horrifying trials to save his Earth from the plague, and to prove that he is the Rightful Heir' 'to say that Mister Monday is definitely worth:I was interested enough in the strange and original concept to read on to the next book, so I would venture to say that Mister Monday is definitely worth reading at least once'
positive	'I recommend THE INTRUDERS if you enjoy:I recommend THE INTRUDERS if you enjoy good writing, but if you want a great story, you should try THE STRAW MEN instead' 'of the major bios on "Big:I've read all of the major bios on "Big Al" and this is by far the best' 'really great fantasy book:this is a really great fantasy book'

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/books-full-bge-aspect",
    "omymble/books-full-bge-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	3	25.1976	78

Label	Training Sample Count
negative	14
neutral	91
positive	62

Training Hyperparameters

batch_size: (64, 64)
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
0.0041	1	0.2476	-
0.2049	50	0.2339	-
0.4098	100	0.2053	-
0.6148	150	0.0231	-
0.8197	200	0.0038	-
1.0246	250	0.0018	-
1.2295	300	0.0017	-
1.4344	350	0.0014	-
1.6393	400	0.0013	-
1.8443	450	0.001	-
2.0492	500	0.001	-
2.2541	550	0.0007	-
2.4590	600	0.0006	-
2.6639	650	0.0007	-
2.8689	700	0.0006	-
3.0738	750	0.0008	-
3.2787	800	0.0007	-
3.4836	850	0.0007	-
3.6885	900	0.0006	-
3.8934	950	0.0006	-
4.0984	1000	0.0007	0.2748
4.3033	1050	0.0009	-
4.5082	1100	0.0006	-
4.7131	1150	0.0006	-
4.9180	1200	0.0005	-

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}
}

omymble
/

books-full-bge-polarity