SetFit Polarity Model with sentence-transformers/paraphrase-mpnet-base-v2

This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses sentence-transformers/paraphrase-mpnet-base-v2 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: sentence-transformers/paraphrase-mpnet-base-v2
Classification head: a LogisticRegression instance
spaCy Model: en_core_web_lg
SetFitABSA Aspect Model: setfit-absa-aspect
SetFitABSA Polarity Model: /private/var/folders/d9/1xz8vq817d3_x9b35qzvvgjc0000gn/T/tmp_tlwkczp/Riyosha/setfit-absa-paraphrase-mpnet-base-v2-MovieReviews-polarity_1
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
positive	'by an amazing score and STELLAR cinematography:Accompanied by an amazing score and STELLAR cinematography, the opening scene perfectly sets you in a mood that simply compels you to not miss a single second of the film, as the story unfolds before your eyes' 'Superbly directed by Masaki Kobayashi:Superbly directed by Masaki Kobayashi, masterful performance from Tatsuya Nakadai, stunning cinematography from Yoshio Miyajima, and an affecting score composed by Toru Takemitsu – all some of the best at what they did.' 'Kobayashi, masterful performance from Tatsuya Nakadai:Superbly directed by Masaki Kobayashi, masterful performance from Tatsuya Nakadai, stunning cinematography from Yoshio Miyajima, and an affecting score composed by Toru Takemitsu – all some of the best at what they did.'
neutral	', as the story unfolds before your:Accompanied by an amazing score and STELLAR cinematography, the opening scene perfectly sets you in a mood that simply compels you to not miss a single second of the film, as the story unfolds before your eyes' "n't think your cast and crew should:But I don't think your cast and crew should be exterminated, as I said they are mostly good and some even tried. You had some great scores and some of the visuals while random and clearly CGI, they weren't as terrible as I expected, and some make for some nice visuals." "your cast and crew should be exterminated:But I don't think your cast and crew should be exterminated, as I said they are mostly good and some even tried. You had some great scores and some of the visuals while random and clearly CGI, they weren't as terrible as I expected, and some make for some nice visuals."
negative	", but the story and characters just:The production and art direction were so well done, but the story and characters just weren't there for me" "the story and characters just weren't:The production and art direction were so well done, but the story and characters just weren't there for me" 'The main actor felt like he:The main actor felt like he was bored.'

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(
    "setfit-absa-aspect",
    "/private/var/folders/d9/1xz8vq817d3_x9b35qzvvgjc0000gn/T/tmp_tlwkczp/Riyosha/setfit-absa-paraphrase-mpnet-base-v2-MovieReviews-polarity_1",
)
# 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	13	32.7	57

Label	Training Sample Count
negative	9
neutral	5
positive	16

Training Hyperparameters

batch_size: (16, 2)
num_epochs: (1, 16)
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: False
warmup_proportion: 0.1
l2_weight: 0.01
seed: 42
eval_max_steps: -1
load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0294	1	0.2041	-

Framework Versions

Python: 3.9.6
SetFit: 1.1.1
Sentence Transformers: 3.4.1
spaCy: 3.7.5
Transformers: 4.48.3
PyTorch: 2.6.0
Datasets: 3.3.0
Tokenizers: 0.21.0

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}
}

Riyosha
/

setfit-absa-paraphrase-mpnet-base-v2-MovieReviews-polarity_1