metadata
base_model: projecte-aina/ST-NLI-ca_paraphrase-multilingual-mpnet-base
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: Aquest text és Varis
- text: Aquest text és Mobiliari Urbà
- text: Aquest text és Velocitat
- text: Aquest text és Parcs i Jardins
- text: Aquest text és Enllumenat
inference: true
SetFit with projecte-aina/ST-NLI-ca_paraphrase-multilingual-mpnet-base
This is a SetFit model that can be used for Text Classification. This SetFit model uses projecte-aina/ST-NLI-ca_paraphrase-multilingual-mpnet-base 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: projecte-aina/ST-NLI-ca_paraphrase-multilingual-mpnet-base
- Classification head: a LogisticRegression instance
- Maximum Sequence Length: 128 tokens
- Number of Classes: 14 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 |
---|---|
0 |
|
1 |
|
2 |
|
3 |
|
4 |
|
5 |
|
6 |
|
7 |
|
8 |
|
9 |
|
10 |
|
11 |
|
12 |
|
13 |
|
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("adriansanz/setfitemotions")
# Run inference
preds = model("Aquest text és Varis")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 4 | 4.2143 | 6 |
Label | Training Sample Count |
---|---|
0 | 10 |
1 | 10 |
2 | 10 |
3 | 10 |
4 | 10 |
5 | 10 |
6 | 10 |
7 | 10 |
8 | 10 |
9 | 10 |
10 | 10 |
11 | 10 |
12 | 10 |
13 | 10 |
Training Hyperparameters
- batch_size: (16, 16)
- num_epochs: (3, 3)
- 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: True
Training Results
Epoch | Step | Training Loss | Validation Loss |
---|---|---|---|
0.0009 | 1 | 0.2021 | - |
0.0439 | 50 | 0.0263 | - |
0.0879 | 100 | 0.0032 | - |
0.1318 | 150 | 0.0015 | - |
0.1757 | 200 | 0.0012 | - |
0.2197 | 250 | 0.0007 | - |
0.2636 | 300 | 0.0008 | - |
0.3076 | 350 | 0.0006 | - |
0.3515 | 400 | 0.0003 | - |
0.3954 | 450 | 0.0003 | - |
0.4394 | 500 | 0.0004 | - |
0.4833 | 550 | 0.0005 | - |
0.5272 | 600 | 0.0004 | - |
0.5712 | 650 | 0.0005 | - |
0.6151 | 700 | 0.0005 | - |
0.6591 | 750 | 0.0002 | - |
0.7030 | 800 | 0.0001 | - |
0.7469 | 850 | 0.0004 | - |
0.7909 | 900 | 0.0002 | - |
0.8348 | 950 | 0.0003 | - |
0.8787 | 1000 | 0.0002 | - |
0.9227 | 1050 | 0.0002 | - |
0.9666 | 1100 | 0.0003 | - |
1.0105 | 1150 | 0.0002 | - |
1.0545 | 1200 | 0.0002 | - |
1.0984 | 1250 | 0.0002 | - |
1.1424 | 1300 | 0.0003 | - |
1.1863 | 1350 | 0.0003 | - |
1.2302 | 1400 | 0.0001 | - |
1.2742 | 1450 | 0.0002 | - |
1.3181 | 1500 | 0.0001 | - |
1.3620 | 1550 | 0.0001 | - |
1.4060 | 1600 | 0.0003 | - |
1.4499 | 1650 | 0.0001 | - |
1.4938 | 1700 | 0.0001 | - |
1.5378 | 1750 | 0.0001 | - |
1.5817 | 1800 | 0.0001 | - |
1.6257 | 1850 | 0.0001 | - |
1.6696 | 1900 | 0.0001 | - |
1.7135 | 1950 | 0.0001 | - |
1.7575 | 2000 | 0.0002 | - |
1.8014 | 2050 | 0.0001 | - |
1.8453 | 2100 | 0.0001 | - |
1.8893 | 2150 | 0.0002 | - |
1.9332 | 2200 | 0.0001 | - |
1.9772 | 2250 | 0.0002 | - |
2.0211 | 2300 | 0.0001 | - |
2.0650 | 2350 | 0.0001 | - |
2.1090 | 2400 | 0.0001 | - |
2.1529 | 2450 | 0.0001 | - |
2.1968 | 2500 | 0.0001 | - |
2.2408 | 2550 | 0.0001 | - |
2.2847 | 2600 | 0.0 | - |
2.3286 | 2650 | 0.0001 | - |
2.3726 | 2700 | 0.0001 | - |
2.4165 | 2750 | 0.0001 | - |
2.4605 | 2800 | 0.0001 | - |
2.5044 | 2850 | 0.0001 | - |
2.5483 | 2900 | 0.0001 | - |
2.5923 | 2950 | 0.0001 | - |
2.6362 | 3000 | 0.0001 | - |
2.6801 | 3050 | 0.0001 | - |
2.7241 | 3100 | 0.0001 | - |
2.7680 | 3150 | 0.0001 | - |
2.8120 | 3200 | 0.0001 | - |
2.8559 | 3250 | 0.0001 | - |
2.8998 | 3300 | 0.0001 | - |
2.9438 | 3350 | 0.0001 | - |
2.9877 | 3400 | 0.0001 | - |
Framework Versions
- Python: 3.10.12
- SetFit: 1.0.3
- Sentence Transformers: 3.0.1
- Transformers: 4.39.0
- PyTorch: 2.3.1+cu121
- Datasets: 2.20.0
- Tokenizers: 0.15.2
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}
}