SetFit with avsolatorio/GIST-small-Embedding-v0

This is a SetFit model that can be used for Text Classification. This SetFit model uses avsolatorio/GIST-small-Embedding-v0 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

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.

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: avsolatorio/GIST-small-Embedding-v0
Classification head: a LogisticRegression instance
Maximum Sequence Length: 512 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
negative	'totally overwrought , deeply biased , and wholly designed to make you feel guilty about ignoring what the filmmakers clearly believe are the greatest musicians of all time .' 'why did they deem it necessary to document all this emotional misery ?' 'a film without surprise geared toward maximum comfort and familiarity .'
positive	'a moving , if uneven , success .' 'as the dominant christine , sylvie testud is icily brilliant .' "it 's a sharp movie about otherwise dull subjects ."

Evaluation

Metrics

Label	Accuracy
all	0.8677

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 SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("setfit_model_id")
# Run inference
preds = model("a film without surprise geared toward maximum comfort and familiarity .")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	16.875	32

Label	Training Sample Count
negative	8
positive	8

Training Hyperparameters

batch_size: (32, 32)
num_epochs: (10, 10)
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
0.2	1	0.2662	-
10.0	50	0.1229	-

Framework Versions

Python: 3.12.2
SetFit: 1.1.0
Sentence Transformers: 3.1.1
Transformers: 4.45.0
PyTorch: 2.4.1+cpu
Datasets: 3.0.1
Tokenizers: 0.20.0

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}
}

zacpt99
/

GIST-small-Embedding-v0-8-shot