SetFit with sentence-transformers/all-mpnet-base-v2
This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-mpnet-base-v2 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: sentence-transformers/all-mpnet-base-v2
- Classification head: a LogisticRegression instance
- Maximum Sequence Length: 384 tokens
- Number of Classes: 9 classes
Model Sources
- Repository: SetFit on GitHub
- Paper: Efficient Few-Shot Learning Without Prompts
- Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts
Model Labels
Lables
0-press release/advertisement/newspaper publication
1-business updates/strategic announcemet/clarification sought
2-Investor meetings/board meeting
3-earnings call transcript
4-esop/esps
5-violation/litigation/penalty
6-auditors report/result
7-research
8-resignation
| Label | Examples |
|---|---|
| 0 |
|
| 1 |
|
| 2 |
|
| 3 |
|
| 4 |
|
| 5 |
|
| 6 |
|
| 7 |
|
| 8 |
|
Evaluation
Metrics
| Label | Accuracy |
|---|---|
| all | 0.9558 |
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("krish2505/setfitmkrt2")
# Run inference
preds = model("Infosys Unveils New AI-Powered Solutions for Enhanced Customer Experience")
Training Details
Training Set Metrics
| Training set | Min | Median | Max |
|---|---|---|---|
| Word count | 1 | 14.7272 | 50 |
| Label | Training Sample Count |
|---|---|
| 0 | 142 |
| 1 | 134 |
| 2 | 298 |
| 3 | 66 |
| 4 | 43 |
| 5 | 53 |
| 6 | 202 |
| 7 | 34 |
| 8 | 36 |
Training Hyperparameters
- batch_size: (64, 64)
- num_epochs: (2, 2)
- max_steps: -1
- sampling_strategy: oversampling
- num_iterations: 20
- body_learning_rate: (2e-05, 2e-05)
- head_learning_rate: 2e-05
- 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 |
|---|---|---|---|
| 0.0016 | 1 | 0.1754 | - |
| 0.0794 | 50 | 0.0917 | - |
| 0.1587 | 100 | 0.0534 | - |
| 0.2381 | 150 | 0.0521 | - |
| 0.3175 | 200 | 0.0352 | - |
| 0.3968 | 250 | 0.0062 | - |
| 0.4762 | 300 | 0.0159 | - |
| 0.5556 | 350 | 0.0151 | - |
| 0.6349 | 400 | 0.0207 | - |
| 0.7143 | 450 | 0.0129 | - |
| 0.7937 | 500 | 0.0186 | - |
| 0.8730 | 550 | 0.0083 | - |
| 0.9524 | 600 | 0.002 | - |
| 1.0317 | 650 | 0.0081 | - |
| 1.1111 | 700 | 0.0263 | - |
| 1.1905 | 750 | 0.0118 | - |
| 1.2698 | 800 | 0.0196 | - |
| 1.3492 | 850 | 0.011 | - |
| 1.4286 | 900 | 0.0153 | - |
| 1.5079 | 950 | 0.0015 | - |
| 1.5873 | 1000 | 0.0156 | - |
| 1.6667 | 1050 | 0.0215 | - |
| 1.7460 | 1100 | 0.0022 | - |
| 1.8254 | 1150 | 0.003 | - |
| 1.9048 | 1200 | 0.0033 | - |
| 1.9841 | 1250 | 0.0155 | - |
Framework Versions
- Python: 3.10.12
- SetFit: 1.0.3
- Sentence Transformers: 2.2.2
- Transformers: 4.36.2
- PyTorch: 2.0.0
- Datasets: 2.16.1
- Tokenizers: 0.15.0
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}
}
- Downloads last month
- 0