Hugging Face's logo

fefofico
/
nuclear_trainer

Text Classification
setfit
Safetensors
sentence-transformers
modernbert
generated_from_setfit_trainer
Eval Results (legacy)
text-embeddings-inference
Model card Files
xet
Community
nuclear_trainer / README.md
fefofico's picture
fefofico
Add SetFit model
0b362e4 verified
preview code
|
raw
history blame contribute delete
20.6 kB
 ---
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: we are determined to reinvigorate our political dialogue including on strategic
issues, such as energy security and counter terrorism.
- text: We are committed to making full use of the potential of existing NATO-Russia
agreements and invite Russia to do likewise.
- text: and the fact that this treaty is now is in jeopardy and that time is running
out for saving the treaty, of course, it’s extremely serious for arms control.
- text: so, this is a modernization of the nuclear deterrent we have for many years.
- text: i just had a professional exchange with minister lavrov, where we covered
a wide range of different issues, including the inf treaty, ukraine, afghanistan,
and also the general need for dialogue nato-russia, which covers issues as risk
reduction, transparency and how to brief each other on, for instance, upcoming
exercises.
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: nomic-ai/modernbert-embed-base
model-index:
- name: SetFit with nomic-ai/modernbert-embed-base
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 0.9168443496801706
name: Accuracy
---
# SetFit with nomic-ai/modernbert-embed-base
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [nomic-ai/modernbert-embed-base](https://huggingface.co/nomic-ai/modernbert-embed-base) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) 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](https://www.sbert.net) 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:** [nomic-ai/modernbert-embed-base](https://huggingface.co/nomic-ai/modernbert-embed-base)
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
- **Maximum Sequence Length:** 8192 tokens
- **Number of Classes:** 2 classes
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
### Model Labels
| Label | Examples |
|:------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| 0 | <ul><li>'there has been significant reconstruction and development, especially in the north of the country, and afghanistan’s gross national product has tripled over the past few years.'</li><li>'a number of commentators wrongly analysed the debate of last february as the end of the alliance.'</li><li>'china has the right to, as all other nations to exercise their forces.'</li></ul> |
| 1 | <ul><li>'but we also need to take into account the security consequence for us here by the rise of china, investing in hypersonic glide vehicles, long range … significantly increasing their nuclear arsenals.'</li><li>'as a first step, we are proposing mutual briefings on exercises and nuclear policies in the nato-russia council.'</li><li>"We underscore that Russia's irresponsible nuclear rhetoric is unacceptable and that any use of nuclear weapons would meet with unequivocal international condemnation and severe consequences."</li></ul> |
## Evaluation
### Metrics
| Label | Accuracy |
|:--------|:---------|
| **all** | 0.9168 |
## Uses
### Direct Use for Inference
First install the SetFit library:
```bash
pip install setfit
```
Then you can load this model and run inference.
```python
from setfit import SetFitModel
# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("fefofico/nuclear_trained")
# Run inference
preds = model("so, this is a modernization of the nuclear deterrent we have for many years.")
```
<!--
### Downstream Use
*List how someone could finetune this model on their own dataset.*
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Set Metrics
| Training set | Min | Median | Max |
|:-------------|:----|:--------|:----|
| Word count | 2 | 24.7149 | 132 |
| Label | Training Sample Count |
|:------|:----------------------|
| 0 | 1017 |
| 1 | 856 |
### Training Hyperparameters
- batch_size: (20, 20)
- num_epochs: (20, 20)
- max_steps: -1
- sampling_strategy: oversampling
- num_iterations: 3
- 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 |
|:-------:|:-----:|:-------------:|:---------------:|
| 0.0018 | 1 | 0.258 | - |
| 0.0890 | 50 | 0.2535 | - |
| 0.1779 | 100 | 0.2445 | - |
| 0.2669 | 150 | 0.2423 | - |
| 0.3559 | 200 | 0.2315 | - |
| 0.4448 | 250 | 0.2077 | - |
| 0.5338 | 300 | 0.1586 | - |
| 0.6228 | 350 | 0.136 | - |
| 0.7117 | 400 | 0.1016 | - |
| 0.8007 | 450 | 0.0879 | - |
| 0.8897 | 500 | 0.0641 | - |
| 0.9786 | 550 | 0.0523 | - |
| 1.0676 | 600 | 0.0456 | - |
| 1.1566 | 650 | 0.0358 | - |
| 1.2456 | 700 | 0.0243 | - |
| 1.3345 | 750 | 0.0197 | - |
| 1.4235 | 800 | 0.0173 | - |
| 1.5125 | 850 | 0.0103 | - |
| 1.6014 | 900 | 0.0105 | - |
| 1.6904 | 950 | 0.0118 | - |
| 1.7794 | 1000 | 0.0202 | - |
| 1.8683 | 1050 | 0.0124 | - |
| 1.9573 | 1100 | 0.0118 | - |
| 2.0463 | 1150 | 0.0074 | - |
| 2.1352 | 1200 | 0.0045 | - |
| 2.2242 | 1250 | 0.0036 | - |
| 2.3132 | 1300 | 0.0068 | - |
| 2.4021 | 1350 | 0.0032 | - |
| 2.4911 | 1400 | 0.0012 | - |
| 2.5801 | 1450 | 0.0021 | - |
| 2.6690 | 1500 | 0.0021 | - |
| 2.7580 | 1550 | 0.0003 | - |
| 2.8470 | 1600 | 0.0025 | - |
| 2.9359 | 1650 | 0.0003 | - |
| 3.0249 | 1700 | 0.0002 | - |
| 3.1139 | 1750 | 0.0002 | - |
| 3.2028 | 1800 | 0.0001 | - |
| 3.2918 | 1850 | 0.0001 | - |
| 3.3808 | 1900 | 0.0001 | - |
| 3.4698 | 1950 | 0.0001 | - |
| 3.5587 | 2000 | 0.0003 | - |
| 3.6477 | 2050 | 0.0001 | - |
| 3.7367 | 2100 | 0.0004 | - |
| 3.8256 | 2150 | 0.0009 | - |
| 3.9146 | 2200 | 0.0001 | - |
| 4.0036 | 2250 | 0.0006 | - |
| 4.0925 | 2300 | 0.0005 | - |
| 4.1815 | 2350 | 0.0001 | - |
| 4.2705 | 2400 | 0.0001 | - |
| 4.3594 | 2450 | 0.0001 | - |
| 4.4484 | 2500 | 0.0001 | - |
| 4.5374 | 2550 | 0.0001 | - |
| 4.6263 | 2600 | 0.0001 | - |
| 4.7153 | 2650 | 0.0001 | - |
| 4.8043 | 2700 | 0.0001 | - |
| 4.8932 | 2750 | 0.0 | - |
| 4.9822 | 2800 | 0.0003 | - |
| 5.0712 | 2850 | 0.0 | - |
| 5.1601 | 2900 | 0.0 | - |
| 5.2491 | 2950 | 0.0 | - |
| 5.3381 | 3000 | 0.0 | - |
| 5.4270 | 3050 | 0.0 | - |
| 5.5160 | 3100 | 0.0 | - |
| 5.6050 | 3150 | 0.0002 | - |
| 5.6940 | 3200 | 0.0 | - |
| 5.7829 | 3250 | 0.0 | - |
| 5.8719 | 3300 | 0.0001 | - |
| 5.9609 | 3350 | 0.0 | - |
| 6.0498 | 3400 | 0.0 | - |
| 6.1388 | 3450 | 0.0 | - |
| 6.2278 | 3500 | 0.0 | - |
| 6.3167 | 3550 | 0.0 | - |
| 6.4057 | 3600 | 0.0 | - |
| 6.4947 | 3650 | 0.0 | - |
| 6.5836 | 3700 | 0.0 | - |
| 6.6726 | 3750 | 0.0 | - |
| 6.7616 | 3800 | 0.0 | - |
| 6.8505 | 3850 | 0.0 | - |
| 6.9395 | 3900 | 0.0 | - |
| 7.0285 | 3950 | 0.0 | - |
| 7.1174 | 4000 | 0.0 | - |
| 7.2064 | 4050 | 0.0 | - |
| 7.2954 | 4100 | 0.0 | - |
| 7.3843 | 4150 | 0.0 | - |
| 7.4733 | 4200 | 0.0 | - |
| 7.5623 | 4250 | 0.0 | - |
| 7.6512 | 4300 | 0.0 | - |
| 7.7402 | 4350 | 0.0 | - |
| 7.8292 | 4400 | 0.0 | - |
| 7.9181 | 4450 | 0.0 | - |
| 8.0071 | 4500 | 0.0 | - |
| 8.0961 | 4550 | 0.0 | - |
| 8.1851 | 4600 | 0.0 | - |
| 8.2740 | 4650 | 0.0 | - |
| 8.3630 | 4700 | 0.0 | - |
| 8.4520 | 4750 | 0.0 | - |
| 8.5409 | 4800 | 0.0 | - |
| 8.6299 | 4850 | 0.0 | - |
| 8.7189 | 4900 | 0.0 | - |
| 8.8078 | 4950 | 0.0 | - |
| 8.8968 | 5000 | 0.0 | - |
| 8.9858 | 5050 | 0.0 | - |
| 9.0747 | 5100 | 0.0 | - |
| 9.1637 | 5150 | 0.0 | - |
| 9.2527 | 5200 | 0.0 | - |
| 9.3416 | 5250 | 0.0 | - |
| 9.4306 | 5300 | 0.0 | - |
| 9.5196 | 5350 | 0.0 | - |
| 9.6085 | 5400 | 0.0 | - |
| 9.6975 | 5450 | 0.0 | - |
| 9.7865 | 5500 | 0.0 | - |
| 9.8754 | 5550 | 0.0 | - |
| 9.9644 | 5600 | 0.0 | - |
| 10.0534 | 5650 | 0.0 | - |
| 10.1423 | 5700 | 0.0 | - |
| 10.2313 | 5750 | 0.0 | - |
| 10.3203 | 5800 | 0.0 | - |
| 10.4093 | 5850 | 0.0 | - |
| 10.4982 | 5900 | 0.0 | - |
| 10.5872 | 5950 | 0.0 | - |
| 10.6762 | 6000 | 0.0 | - |
| 10.7651 | 6050 | 0.0 | - |
| 10.8541 | 6100 | 0.0 | - |
| 10.9431 | 6150 | 0.0 | - |
| 11.0320 | 6200 | 0.0 | - |
| 11.1210 | 6250 | 0.0 | - |
| 11.2100 | 6300 | 0.0 | - |
| 11.2989 | 6350 | 0.0 | - |
| 11.3879 | 6400 | 0.0 | - |
| 11.4769 | 6450 | 0.0 | - |
| 11.5658 | 6500 | 0.0 | - |
| 11.6548 | 6550 | 0.0 | - |
| 11.7438 | 6600 | 0.0 | - |
| 11.8327 | 6650 | 0.0 | - |
| 11.9217 | 6700 | 0.0 | - |
| 12.0107 | 6750 | 0.0 | - |
| 12.0996 | 6800 | 0.0 | - |
| 12.1886 | 6850 | 0.0 | - |
| 12.2776 | 6900 | 0.0 | - |
| 12.3665 | 6950 | 0.0 | - |
| 12.4555 | 7000 | 0.0 | - |
| 12.5445 | 7050 | 0.0 | - |
| 12.6335 | 7100 | 0.0 | - |
| 12.7224 | 7150 | 0.0 | - |
| 12.8114 | 7200 | 0.0 | - |
| 12.9004 | 7250 | 0.0 | - |
| 12.9893 | 7300 | 0.0 | - |
| 13.0783 | 7350 | 0.0 | - |
| 13.1673 | 7400 | 0.0 | - |
| 13.2562 | 7450 | 0.0 | - |
| 13.3452 | 7500 | 0.0 | - |
| 13.4342 | 7550 | 0.0 | - |
| 13.5231 | 7600 | 0.0 | - |
| 13.6121 | 7650 | 0.0 | - |
| 13.7011 | 7700 | 0.0 | - |
| 13.7900 | 7750 | 0.0 | - |
| 13.8790 | 7800 | 0.0 | - |
| 13.9680 | 7850 | 0.0 | - |
| 14.0569 | 7900 | 0.0 | - |
| 14.1459 | 7950 | 0.0 | - |
| 14.2349 | 8000 | 0.0 | - |
| 14.3238 | 8050 | 0.0 | - |
| 14.4128 | 8100 | 0.0 | - |
| 14.5018 | 8150 | 0.0 | - |
| 14.5907 | 8200 | 0.0 | - |
| 14.6797 | 8250 | 0.0 | - |
| 14.7687 | 8300 | 0.0 | - |
| 14.8577 | 8350 | 0.0 | - |
| 14.9466 | 8400 | 0.0 | - |
| 15.0356 | 8450 | 0.0 | - |
| 15.1246 | 8500 | 0.0 | - |
| 15.2135 | 8550 | 0.0 | - |
| 15.3025 | 8600 | 0.0 | - |
| 15.3915 | 8650 | 0.0 | - |
| 15.4804 | 8700 | 0.0 | - |
| 15.5694 | 8750 | 0.0 | - |
| 15.6584 | 8800 | 0.0 | - |
| 15.7473 | 8850 | 0.0 | - |
| 15.8363 | 8900 | 0.0 | - |
| 15.9253 | 8950 | 0.0 | - |
| 16.0142 | 9000 | 0.0 | - |
| 16.1032 | 9050 | 0.0 | - |
| 16.1922 | 9100 | 0.0 | - |
| 16.2811 | 9150 | 0.0 | - |
| 16.3701 | 9200 | 0.0 | - |
| 16.4591 | 9250 | 0.0 | - |
| 16.5480 | 9300 | 0.0 | - |
| 16.6370 | 9350 | 0.0 | - |
| 16.7260 | 9400 | 0.0 | - |
| 16.8149 | 9450 | 0.0 | - |
| 16.9039 | 9500 | 0.0 | - |
| 16.9929 | 9550 | 0.0 | - |
| 17.0819 | 9600 | 0.0 | - |
| 17.1708 | 9650 | 0.0 | - |
| 17.2598 | 9700 | 0.0 | - |
| 17.3488 | 9750 | 0.0 | - |
| 17.4377 | 9800 | 0.0 | - |
| 17.5267 | 9850 | 0.0 | - |
| 17.6157 | 9900 | 0.0 | - |
| 17.7046 | 9950 | 0.0 | - |
| 17.7936 | 10000 | 0.0 | - |
| 17.8826 | 10050 | 0.0 | - |
| 17.9715 | 10100 | 0.0 | - |
| 18.0605 | 10150 | 0.0 | - |
| 18.1495 | 10200 | 0.0 | - |
| 18.2384 | 10250 | 0.0 | - |
| 18.3274 | 10300 | 0.0 | - |
| 18.4164 | 10350 | 0.0 | - |
| 18.5053 | 10400 | 0.0 | - |
| 18.5943 | 10450 | 0.0 | - |
| 18.6833 | 10500 | 0.0 | - |
| 18.7722 | 10550 | 0.0 | - |
| 18.8612 | 10600 | 0.0 | - |
| 18.9502 | 10650 | 0.0 | - |
| 19.0391 | 10700 | 0.0 | - |
| 19.1281 | 10750 | 0.0 | - |
| 19.2171 | 10800 | 0.0 | - |
| 19.3060 | 10850 | 0.0 | - |
| 19.3950 | 10900 | 0.0 | - |
| 19.4840 | 10950 | 0.0 | - |
| 19.5730 | 11000 | 0.0 | - |
| 19.6619 | 11050 | 0.0 | - |
| 19.7509 | 11100 | 0.0 | - |
| 19.8399 | 11150 | 0.0 | - |
| 19.9288 | 11200 | 0.0 | - |
### Framework Versions
- Python: 3.12.12
- SetFit: 1.1.3
- Sentence Transformers: 5.1.2
- Transformers: 4.57.1
- PyTorch: 2.8.0+cu126
- Datasets: 4.0.0
- Tokenizers: 0.22.1
## Citation
### BibTeX
```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}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->