Hugging Face's logo

cnmoro
/
prompt-router

Text Classification
setfit
Safetensors
sentence-transformers
xlm-roberta
generated_from_setfit_trainer
Eval Results
Model card Files
xet
Community
prompt-router / README.md
cnmoro's picture
cnmoro
Update README.md
7ec283f verified
preview code
|
raw
history blame contribute delete
24.5 kB
 ---
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: A fantastical portal opening into another dimension, swirling energy.
- text: Analyze the concept of political trust and its importance for governance.
- text: What makes a particular escape room experience engaging and successful?
- text: What is the function of the lymphatic system?
- text: Desenvolva um conto fictício sobre um mapa antigo que guia para um tesouro
cultural perdido.
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: ibm-granite/granite-embedding-107m-multilingual
model-index:
- name: SetFit with ibm-granite/granite-embedding-107m-multilingual
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 0.8924137931034483
name: Accuracy
---
As of 28/07/2025, I instead of using this model, a simpler approach would be to just use one of these [Gliclass Models](https://huggingface.co/cnmoro/gliclass-base-v3.0-onnx), matching the user's prompt against the prompts classes. But this model will remain here nonetheless.
------------------------
# SetFit with ibm-granite/granite-embedding-107m-multilingual
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual) 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:** [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual)
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
- **Maximum Sequence Length:** 512 tokens
- **Number of Classes:** 30 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 |
|:-------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| sentiment_analysis | <ul><li>'Como a análise de sentimento pode melhorar a tomada de decisões estratégicas?'</li><li>'Dê um exemplo de como a ironia afeta a análise de sentimento.'</li><li>'What are the key advantages of using transformer-based models (e.g., BERT, RoBERTa) for sentiment analysis tasks?'</li></ul> |
| marketing | <ul><li>'What are the emerging trends in voice search optimization for marketing?'</li><li>'How can augmented reality (AR) be integrated into marketing experiences?'</li><li>'How does psychographic segmentation differ from demographic segmentation?'</li></ul> |
| entertainment | <ul><li>'Explore the challenges of balancing artistic integrity with commercial viability in entertainment.'</li><li>'Discuss the impact of major sporting events as a form of entertainment.'</li><li>"O que torna uma canção um 'hit global' na era da internet?"</li></ul> |
| image_generation | <ul><li>'Um grupo de pássaros migratórios voando em formação perfeita no céu azul, com nuvens ao fundo.'</li><li>'A photorealistic rendering of a gourmet dish, top-down view, professional food photography.'</li><li>'Uma ponte suspensa de madeira em uma floresta tropical densa, com névoa subindo do chão.'</li></ul> |
| complex_reasoning | <ul><li>'Implement a neural-symbolic system that combines rule-based reasoning with deep learning to answer complex scientific questions from literature.'</li><li>'Create a method to reconstruct 3D objects from multiple 2D images taken under varying lighting and viewpoints with occlusions.'</li><li>'Descreva métodos para criar jogos educacionais que se adaptem dinamicamente ao progresso do aluno.'</li></ul> |
| education | <ul><li>'Propose ways to reduce the achievement gap among different socio-economic groups.'</li><li>'O papel do esporte no desenvolvimento integral dos estudantes.'</li><li>'A importância da colaboração entre escolas e comunidade.'</li></ul> |
| mathematics | <ul><li>'Discuss the concept of convexity in optimization.'</li><li>'How does game theory use mathematical models?'</li><li>'Descreva uma estratégia eficaz para resolver problemas de matemática que envolvem múltiplas etapas.'</li></ul> |
| biology | <ul><li>'Explique a regulação da temperatura corporal em humanos.'</li><li>'Descreva os diferentes níveis de organização biológica.'</li><li>'O que é a homeostase e por que ela é vital?'</li></ul> |
| extraction | <ul><li>'Faça uma síntese das principais características dos dados apresentados no relatório de desmatamento, destacando tendências e padrões observáveis.'</li><li>'É seu dever revelar os elementos-chave que explicam a relação entre desmatamento e políticas públicas, destacando causas e soluções propostas.'</li><li>"Build a detailed analysis of the competitor's marketing funnel, from awareness to conversion."</li></ul> |
| engineering | <ul><li>'What are the main methods of controlled demolition of structures?'</li><li>'Como a sustentabilidade pode ser integrada no projeto e construção de edifícios residenciais?'</li><li>'Discuss the application of artificial intelligence in predictive maintenance of electrical equipment.'</li></ul> |
| ethics | <ul><li>'Defina ética e moralidade, destacando suas principais diferenças e interconexões.'</li><li>'Is there a universal ethic that applies to all humans?'</li><li>'How do personal values shape ethical choices?'</li></ul> |
| law | <ul><li>'O que é a Lei Anticorrupção?'</li><li>'Quais os direitos dos animais no direito brasileiro?'</li><li>'Explain the concept of intellectual property.'</li></ul> |
| general_knowledge | <ul><li>'Qual é a importância da agricultura para a economia brasileira?'</li><li>'Quais espécies animais são consideradas ameaçadas de extinção no Brasil?'</li><li>'O que é a imunidade vacinal e como as vacinas funcionam?'</li></ul> |
| geopolitics | <ul><li>'Discuss the role of the International Criminal Court in global justice and accountability.'</li><li>'Examine the role of proxy conflicts in modern geopolitical competition.'</li><li>'Examine the role of the World Trade Organization in a protectionist global economy.'</li></ul> |
| summarization | <ul><li>'Resuma os resultados de uma avaliação educacional nacional.'</li><li>'Resuma um artigo jornalístico investigativo explicando os fatos.'</li><li>'Summarize user experience testing results to prioritize UI improvements.'</li></ul> |
| healthcare | <ul><li>'Describe the role of a paramedic in the pre-hospital emergency care setting.'</li><li>'What is sepsis and why is it a medical emergency?'</li><li>'Discuss the medical implications of an aging global population.'</li></ul> |
| spiritual | <ul><li>'A experiência do êxtase espiritual.'</li><li>'Aspectos do misticismo e o inexplicável.'</li><li>'How do you manage expectations on your spiritual journey?'</li></ul> |
| coding | <ul><li>'Implemente uma função para balancear expressões matemáticas adicionando parênteses corretamente.'</li><li>'Desenvolva um algoritmo que transforme uma expressão regular em um autômato finito determinístico.'</li><li>'Desenvolva um algoritmo para reconhecimento de padrões em strings baseado em autômatos finitos não determinísticos.'</li></ul> |
| tool | <ul><li>'Leia o conteúdo de uma página web e resuma os principais pontos.'</li><li>'Traduza este texto do português para inglês usando um serviço externo.'</li><li>'Retrieve the top 5 upcoming tech conferences worldwide this year.'</li></ul> |
| politics | <ul><li>'Explain the process of judicial review and its role in a constitutional government.'</li><li>'Como as campanhas eleitorais influenciam o eleitorado e quais estratégias são utilizadas?'</li><li>'Discuta o funcionamento de um regime presidencialista e suas vantagens e desvantagens.'</li></ul> |
| business | <ul><li>'Princípios e metodologias da gestão ágil de projetos (Agile) aplicadas a empresas.'</li><li>'O papel do CEO moderno em um cenário de negócios em constante mudança.'</li><li>'How can businesses optimize their operational processes for greater efficiency?'</li></ul> |
| creativity | <ul><li>'Write a story about a labyrinth that reconfigures itself based on the visitor’s fears.'</li><li>'Escreva uma crônica de humor sobre as peculiaridades do transporte público em uma capital brasileira.'</li><li>'Desenvolva um conto que envolva um segredo escondido dentro de uma música popular brasileira.'</li></ul> |
| physics | <ul><li>'What is superconductivity?'</li><li>'O que é a física do plasma e onde ela é estudada/aplicada?'</li><li>'Explain the concept of a quantum field.'</li></ul> |
| psychological | <ul><li>'Identificando e apoiando dificuldades de aprendizagem.'</li><li>'Analyze the concept of social loafing and ways to mitigate it.'</li><li>'Discuss the psychological factors influencing academic performance and learning.'</li></ul> |
| history | <ul><li>'Analyze the concept of "historical turning points."'</li><li>'Explore the history of human-animal relationships.'</li><li>'How did the fall of the Berlin Wall affect European integration?'</li></ul> |
| translation | <ul><li>"Convert this folk song from Portuguese to English: 'Asa Branca - Luiz Gonzaga'."</li><li>"Poderia traduzir esta citação filosófica do latim para português: 'Cogito, ergo sum'."</li><li>"Convert this Brazilian lullaby to English: 'Boi da cara preta, pega essa criança que tem medo de careta.'"</li></ul> |
| basic_reasoning | <ul><li>'Se um carro gasta 10 litros de combustível para percorrer 100 km, quanto gastará em 250 km?'</li><li>'If the sum of two numbers is 35 and their difference is 5, what are the numbers?'</li><li>'O que é maior: 1/2 ou 0,6?'</li></ul> |
| finance | <ul><li>'Describe the process of a company going public (IPO).'</li><li>'Discuss the role of regulations in preventing financial crises.'</li><li>"Discuss the concept of 'too big to fail' in the banking sector."</li></ul> |
| chemistry | <ul><li>'Como a temperatura afeta a velocidade das reações químicas?'</li><li>'O que são os ligantes em compostos de coordenação?'</li><li>'What is green chemistry? List and explain at least three of its core principles.'</li></ul> |
| roleplay | <ul><li>'Personifique um cineasta independendente buscando financiamento para um projeto arriscado.'</li><li>'Atue como um editor de jogos digitais assistindo testes beta e tomando decisões de ajustes finais.'</li><li>'You are a psychologist exploring childhood trauma with a patient using therapeutic techniques.'</li></ul> |
## Evaluation
### Metrics
| Label | Accuracy |
|:--------|:---------|
| **all** | 0.8924 |
## 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("cnmoro/prompt-router")
# Run inference
preds = model("What is the function of the lymphatic system?")
```
<!--
### 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 | 3 | 11.6859 | 38 |
| Label | Training Sample Count |
|:-------------------|:----------------------|
| creativity | 176 |
| extraction | 283 |
| image_generation | 173 |
| education | 181 |
| summarization | 174 |
| chemistry | 174 |
| sentiment_analysis | 179 |
| geopolitics | 181 |
| translation | 179 |
| history | 177 |
| coding | 158 |
| politics | 181 |
| healthcare | 178 |
| business | 170 |
| complex_reasoning | 152 |
| psychological | 174 |
| biology | 172 |
| mathematics | 178 |
| marketing | 177 |
| physics | 177 |
| engineering | 176 |
| roleplay | 171 |
| finance | 175 |
| basic_reasoning | 154 |
| ethics | 180 |
| entertainment | 180 |
| tool | 166 |
| law | 173 |
| spiritual | 175 |
| general_knowledge | 170 |
### Training Hyperparameters
- batch_size: (8, 8)
- num_epochs: (1, 16)
- max_steps: 2400
- 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
- evaluation_strategy: steps
- eval_max_steps: -1
- load_best_model_at_end: True
### Training Results
| Epoch | Step | Training Loss | Validation Loss |
|:------:|:----:|:-------------:|:---------------:|
| 0.0004 | 1 | 0.2374 | - |
| 0.0208 | 50 | 0.2111 | - |
| 0.0417 | 100 | 0.2087 | - |
| 0.0625 | 150 | 0.1995 | - |
| 0.0833 | 200 | 0.1984 | 0.1876 |
| 0.1042 | 250 | 0.1894 | - |
| 0.125 | 300 | 0.1872 | - |
| 0.1458 | 350 | 0.1818 | - |
| 0.1667 | 400 | 0.1758 | 0.1587 |
| 0.1875 | 450 | 0.1647 | - |
| 0.2083 | 500 | 0.1547 | - |
| 0.2292 | 550 | 0.1404 | - |
| 0.25 | 600 | 0.1342 | 0.1252 |
| 0.2708 | 650 | 0.1309 | - |
| 0.2917 | 700 | 0.1209 | - |
| 0.3125 | 750 | 0.1329 | - |
| 0.3333 | 800 | 0.1068 | 0.1055 |
| 0.3542 | 850 | 0.1131 | - |
| 0.375 | 900 | 0.1006 | - |
| 0.3958 | 950 | 0.1033 | - |
| 0.4167 | 1000 | 0.1005 | 0.0922 |
| 0.4375 | 1050 | 0.1133 | - |
| 0.4583 | 1100 | 0.0898 | - |
| 0.4792 | 1150 | 0.0918 | - |
| 0.5 | 1200 | 0.0983 | 0.0855 |
| 0.5208 | 1250 | 0.0947 | - |
| 0.5417 | 1300 | 0.0921 | - |
| 0.5625 | 1350 | 0.1045 | - |
| 0.5833 | 1400 | 0.09 | 0.0763 |
| 0.6042 | 1450 | 0.0893 | - |
| 0.625 | 1500 | 0.0823 | - |
| 0.6458 | 1550 | 0.0853 | - |
| 0.6667 | 1600 | 0.0881 | 0.0713 |
| 0.6875 | 1650 | 0.0837 | - |
| 0.7083 | 1700 | 0.0886 | - |
| 0.7292 | 1750 | 0.0784 | - |
| 0.75 | 1800 | 0.0838 | 0.0680 |
| 0.7708 | 1850 | 0.0743 | - |
| 0.7917 | 1900 | 0.0788 | - |
| 0.8125 | 1950 | 0.084 | - |
| 0.8333 | 2000 | 0.0772 | 0.0659 |
| 0.8542 | 2050 | 0.0872 | - |
| 0.875 | 2100 | 0.0808 | - |
| 0.8958 | 2150 | 0.0649 | - |
| 0.9167 | 2200 | 0.0795 | 0.0651 |
| 0.9375 | 2250 | 0.0774 | - |
| 0.9583 | 2300 | 0.0687 | - |
| 0.9792 | 2350 | 0.0787 | - |
| 1.0 | 2400 | 0.0786 | 0.0647 |
### Framework Versions
- Python: 3.11.11
- SetFit: 1.2.0.dev0
- Sentence Transformers: 5.0.0
- Transformers: 4.53.2
- PyTorch: 2.7.1+cu126
- Datasets: 3.2.0
- Tokenizers: 0.21.0
## 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.*
-->