metadata
library_name: setfit
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
metrics:
- accuracy
widget:
- text: frais douane import vehicule usa carte usd commission
- text: debit automatique assurance chien fido protect
- text: facture carte du cabinet architecte plan maison est carte
- text: achat le monde des oiseaux carte
- text: abonnement mensuel salle de sport life fitness club carte
pipeline_tag: text-classification
inference: true
model-index:
- name: SetFit
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 0.18181818181818182
name: Accuracy
SetFit
This is a SetFit model that can be used for Text Classification. 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
- Classification head: a LogisticRegression instance
- Maximum Sequence Length: 128 tokens
- Number of Classes: 44 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 |
---|---|
Shopping / electronics & multimedia |
|
Other / kids |
|
Bank services / other |
|
Housing / rent |
|
Transportation / other |
|
Bank services / transfers |
|
Investment / retirement & savings |
|
Other / taxes |
|
Healthy & Beauty / other |
|
Investment / securities |
|
Housing / other |
|
Housing / house loan |
|
Housing / utilities & bills |
|
Bank services / general fees |
|
Leisure & Entertainment / culture & events |
|
Transportation / taxi & carpool |
|
Shopping / other |
|
Recurrent Payments / loans |
|
Healthy & Beauty / doctor fees |
|
Bank services / withdrawal |
|
Other / other |
|
Healthy & Beauty / pharmacy |
|
Transportation / fuel |
|
Shopping / sporting goods |
|
Food & Drinks / groceries |
|
Other / pets |
|
Investment / real estate |
|
Shopping / clothing |
|
Shopping / housing equipment |
|
Transportation / maitenance |
|
Recurrent Payments / other |
|
Recurrent Payments / insurance |
|
Healthy & Beauty / veterinary |
|
Transportation / public transportation |
|
Healthy & Beauty / beauty & self-care |
|
Leisure & Entertainment / other |
|
Food & Drinks / eating out |
|
Housing / services & maintenance |
|
Leisure & Entertainment / travel |
|
Leisure & Entertainment / sports & hobbies |
|
Investment / other |
|
Transportation / car loan & leasing |
|
Recurrent Payments / subscription |
|
Food & Drinks / other |
|
Evaluation
Metrics
Label | Accuracy |
---|---|
all | 0.1818 |
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("HEN10/setfit-particular-transaction-solon-embeddings-labels-large-kaggle-automatisation-v1")
# Run inference
preds = model("achat le monde des oiseaux carte")
Training Details
Training Set Metrics
Training set | Min | Median | Max |
---|---|---|---|
Word count | 3 | 6.2727 | 11 |
Label | Training Sample Count |
---|---|
Housing / rent | 2 |
Housing / house loan | 2 |
Housing / utilities & bills | 2 |
Housing / services & maintenance | 2 |
Housing / other | 2 |
Food & Drinks / groceries | 2 |
Food & Drinks / eating out | 2 |
Food & Drinks / other | 2 |
Leisure & Entertainment / sports & hobbies | 2 |
Leisure & Entertainment / culture & events | 2 |
Leisure & Entertainment / travel | 2 |
Leisure & Entertainment / other | 2 |
Transportation / car loan & leasing | 2 |
Transportation / fuel | 2 |
Transportation / public transportation | 2 |
Transportation / taxi & carpool | 2 |
Transportation / maitenance | 2 |
Transportation / other | 2 |
Recurrent Payments / loans | 2 |
Recurrent Payments / insurance | 2 |
Recurrent Payments / subscription | 2 |
Recurrent Payments / other | 2 |
Investment / securities | 2 |
Investment / retirement & savings | 2 |
Investment / real estate | 2 |
Investment / other | 2 |
Shopping / clothing | 2 |
Shopping / electronics & multimedia | 2 |
Shopping / sporting goods | 2 |
Shopping / housing equipment | 2 |
Shopping / other | 2 |
Healthy & Beauty / doctor fees | 2 |
Healthy & Beauty / pharmacy | 2 |
Healthy & Beauty / beauty & self-care | 2 |
Healthy & Beauty / veterinary | 2 |
Healthy & Beauty / other | 2 |
Bank services / transfers | 2 |
Bank services / withdrawal | 2 |
Bank services / general fees | 2 |
Bank services / other | 2 |
Other / taxes | 2 |
Other / kids | 2 |
Other / pets | 2 |
Other / other | 2 |
Training Hyperparameters
- batch_size: (16, 16)
- 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: True
- use_amp: False
- warmup_proportion: 0.1
- seed: 6
- eval_max_steps: -1
- load_best_model_at_end: False
Training Results
Epoch | Step | Training Loss | Validation Loss |
---|---|---|---|
0.0021 | 1 | 0.1221 | - |
0.1057 | 50 | 0.1337 | - |
0.2114 | 100 | 0.0617 | - |
0.3171 | 150 | 0.0397 | - |
0.4228 | 200 | 0.0377 | - |
0.5285 | 250 | 0.0133 | - |
0.6342 | 300 | 0.012 | - |
0.7400 | 350 | 0.0191 | - |
0.8457 | 400 | 0.0118 | - |
0.9514 | 450 | 0.0083 | - |
Framework Versions
- Python: 3.10.13
- SetFit: 1.0.3
- Sentence Transformers: 2.6.1
- Transformers: 4.39.3
- PyTorch: 2.1.2+cpu
- Datasets: 2.17.0
- 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}
}