absa-laptops-aspect / README.md

najwaa

Add SetFit ABSA model

a89d6a9 verified 8 months ago

preview code

raw

history blame contribute delete

8.29 kB

metadata

tags:
  - setfit
  - absa
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
widget:
  - text: >-
      fast:fast boot up, great 1080p resolution, expandable (added 4gb
      additional ram and a 1tb hd) and great value for it's $365+tax price
      point.
  - text: >-
      thinness:the sleekness and thinness of this laptop is lightweight and easy
      to carry.
  - text: >-
      read:when what i heave read, the memory is not upgradeable since it's
      soldered to the board.
  - text: >-
      memory:a good amount of memory. it doesnt need to have a bunch of memory,
      but a decent amount is perfect!
  - text: >-
      wifi:nevertheless great processor, great graphics, 16 gb memory runs cool
      in daily use, battery lasts about 6-7 hours using wifi and video.
metrics:
  - accuracy
pipeline_tag: text-classification
library_name: setfit
inference: false
base_model: sentence-transformers/all-MiniLM-L6-v2

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:

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 this SetFit model to filter these possible aspect span candidates.
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
SetFitABSA Polarity Model: najwaa/absa-laptops-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	'lightweight:very lightweight.' 'carry:this computer is so light weight and easy to carry.' "lightweight:it's lightweight, the screen is decently bright, and it'll go for hours without needing a charge"
no aspect	'computer:this computer is so light weight and easy to carry.' 'weight:this computer is so light weight and easy to carry.' "screen:it's lightweight, the screen is decently bright, and it'll go for hours without needing a charge"

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",
    "najwaa/absa-laptops-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	1	15.9156	37

Label	Training Sample Count
no aspect	251
aspect	140

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
l2_weight: 0.01
seed: 42
eval_max_steps: -1
load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0015	1	0.29	-
0.0770	50	0.2977	0.2646
0.1541	100	0.2622	0.2558
0.2311	150	0.2493	0.2482
0.3082	200	0.2347	0.2261
0.3852	250	0.1396	0.1701
0.4622	300	0.0514	0.1434
0.5393	350	0.0227	0.1808
0.6163	400	0.0161	0.1624
0.6934	450	0.011	0.1718
0.7704	500	0.0101	0.1731
0.8475	550	0.0089	0.1433
0.9245	600	0.0061	0.1682
1.0015	650	0.0086	0.1627
1.0786	700	0.0078	0.1767
1.1556	750	0.0068	0.1773
1.2327	800	0.0065	0.1766

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}
}