Edit model card

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

This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses sentence-transformers/all-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:

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

Model Details

Model Description

Model Sources

Model Labels

Label Examples
neutral
  • 'skip taking the cord with me because:I charge it at night and skip taking the cord with me because of the good battery life.'
  • 'The tech guy then said the: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.'
  • 'all dark, power light steady, hard:\xa0One night I turned the freaking thing off after using it, the next day I turn it on, no GUI, screen all dark, power light steady, hard drive light steady and not flashing as it usually does.'
positive
  • 'of the good battery life.:I charge it at night and skip taking the cord with me because of the good battery life.'
  • 'is of high quality, has a:it is of high quality, has a killer GUI, is extremely stable, is highly expandable, is bundled with lots of very good applications, is easy to use, and is absolutely gorgeous.'
  • 'has a killer GUI, is extremely:it is of high quality, has a killer GUI, is extremely stable, is highly expandable, is bundled with lots of very good applications, is easy to use, and is absolutely gorgeous.'
negative
  • 'then said the service center does not do: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.'
  • 'concern to the "sales" team, which is: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.'
  • 'on, no GUI, screen all:\xa0One night I turned the freaking thing off after using it, the next day I turn it on, no GUI, screen all dark, power light steady, hard drive light steady and not flashing as it usually does.'
conflict
  • '-No backlit keyboard, but not:-No backlit keyboard, but not an issue for me.'
  • "to replace the battery once, but:I did have to replace the battery once, but that was only a couple months ago and it's been working perfect ever since."

Evaluation

Metrics

Label Accuracy
all 0.7008

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 3 25.5873 48
Label Training Sample Count
conflict 2
negative 45
neutral 30
positive 49

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
0.0120 1 0.2721 -
0.6024 50 0.0894 0.2059
1.2048 100 0.0014 0.2309
1.8072 150 0.0006 0.2359
2.4096 200 0.0005 0.2373
3.0120 250 0.0004 0.2364
3.6145 300 0.0003 0.2371
  • 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}
}
Downloads last month
48
Safetensors
Model size
109M params
Tensor type
F32
·
Inference Examples
Inference API (serverless) has been turned off for this model.

Model tree for joshuasundance/setfit-absa-all-mpnet-base-v2-laptops-polarity

Finetuned
(152)
this model

Evaluation results