ANGKJ1995/GIST-small-Embedding-v0-checkthat-fitset-128

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:

1. Fine-tuning a Sentence Transformer with contrastive learning.
2. 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
subjective	'Stakeholder capitalism poisons democracy and partisan politics poisons capitalism.' 'There is yet everywhere a deficit in the public revenue because the shrinkage in everything taxable was so sudden and violent.' 'Our system of unbridled profit-focused capitalism used to serve as perhaps the most important of those sanctuaries, but no longer.'
objective	'But a top buying agent tells me that access to 13 can be gained if you know the right people.' 'A portion of positive tests around the country is being forwarded to the agency for genetic sequencing, according to a report by CBS News.' 'asked American Federation of Teachers President Randi Weingarten.'

Evaluation

Metrics

Label	Accuracy
all	0.8446

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("As the total national income falls, the proportion of it absorbed by government will rise.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	1	22.9219	77

Label	Training Sample Count
objective	128
subjective	128

Training Hyperparameters

• batch_size: (32, 32)
• num_epochs: (1, 1)
• 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
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0010	1	0.2715	-
0.0484	50	0.2469	-
0.0969	100	0.2247	-
0.1453	150	0.0501	-
0.1938	200	0.0039	-
0.2422	250	0.0014	-
0.2907	300	0.0011	-
0.3391	350	0.0014	-
0.3876	400	0.001	-
0.4360	450	0.0009	-
0.4845	500	0.0008	-
0.5329	550	0.0008	-
0.5814	600	0.0008	-
0.6298	650	0.0007	-
0.6783	700	0.0007	-
0.7267	750	0.0006	-
0.7752	800	0.0007	-
0.8236	850	0.0006	-
0.8721	900	0.0005	-
0.9205	950	0.0007	-
0.9690	1000	0.0007	-

Framework Versions

• Python: 3.11.9
• SetFit: 1.0.3
• Sentence Transformers: 3.0.0
• Transformers: 4.40.2
• PyTorch: 2.1.2
• Datasets: 2.19.1
• Tokenizers: 0.19.1

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}
}