---
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 'Thesis: In my opinion, watching sports on TV is a good opportunity to get
    relax. There are many reasons why some people like to watch sport games on TV.
    Last argument: None Target sentence: Where is the truth?'
- text: 'Thesis: As for me, I suppose that watching sports via TV won''t help you
    become a professional at sport. Moreover, sport events are not very interesting
    to be seen through television channels. Last argument: Besides, it''s much more
    attractive to visit the play in the real life than to watch it at home for many
    reasons. Target sentence: If you watch the sport programme, you can see only those
    things that the operator wants to record.'
- text: 'Thesis: Today there are people who belived that watching any sport is a useless
    time spent. I complitely disagree with this opinion. Last argument: Watching any
    sort games or individual competition is wonderfull way to spend your free time,
    by this hobby you can have a lot of profits. Target sentence: Watching any sort
    games or individual competition is wonderfull way to spend your free time, by
    this hobby you can have a lot of profits.'
- text: 'Thesis: As for me, I suppose that watching sports via TV won''t help you
    become a professional at sport. Moreover, sport events are not very interesting
    to be seen through television channels. Last argument: Besides, it''s much more
    attractive to visit the play in the real life than to watch it at home for many
    reasons. Target sentence: Besides, it''s much more attractive to visit the play
    in the real life than to watch it at home for many reasons.'
- text: 'Thesis: Some people consider that it is a waste of time, bit I desagree with
    this statment and try to refute it. Last argument: If looks on this statement
    otherwise, it is importnat to say that watching sport events is a usual hobby
    such as cooking, reading books and others. Target sentence: People who work in
    public catering like cooking and it is their hobby maybe.'
metrics:
- accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
datasets:
- Zlovoblachko/DeepSeek_dim1
base_model: BAAI/bge-small-en-v1.5
model-index:
- name: SetFit with BAAI/bge-small-en-v1.5
  results:
  - task:
      type: text-classification
      name: Text Classification
    dataset:
      name: Zlovoblachko/DeepSeek_dim1
      type: Zlovoblachko/DeepSeek_dim1
      split: test
    metrics:
    - type: accuracy
      value: 0.8740740740740741
      name: Accuracy
---

# SetFit with BAAI/bge-small-en-v1.5

This is a [SetFit](https://github.com/huggingface/setfit) model trained on the [Zlovoblachko/DeepSeek_dim1](https://huggingface.co/datasets/Zlovoblachko/DeepSeek_dim1) dataset that can be used for Text Classification. This SetFit model uses [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) 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:** [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5)
- **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:** 2 classes
- **Training Dataset:** [Zlovoblachko/DeepSeek_dim1](https://huggingface.co/datasets/Zlovoblachko/DeepSeek_dim1)
<!-- - **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                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
|:------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| L     | <ul><li>'Thesis:  in my opinion it is uself and fun to do pysical exercise and my activity every day. I disagre wits his opinion because these people do not understand that sport should keep fit and mind. Last argument:  Secondly the sport is very fun. Target sentence: However some people prefer to watch sports show on tv.'</li><li>'Thesis:  I personally disadree with this opinion because there are many reasons why watching sports can be useful for people. Last argument: However, I can understend people who agree with first point of view. Target sentence: On the other hand, sometimes it is very difficult to control the time which they lose for that.'</li><li>"Thesis: None Last argument:  People opposing this position may say that sport is not an intellectual activity, so it is not worth spending time at all. Target sentence:  They think that watching sport is not developing people's mind, so it has no sense to watch such things."</li></ul>                                                                                                    |
| H     | <ul><li>'Thesis:  I think, that watching sports really gives a lot of fun. Last argument: The second point is that sports fans are fond of their teams ond sports favoritres. Target sentence: Sports stimulates them to travel (like UFC or Olympic games), to collect merch, to have new datings.'</li><li>'Thesis: So, can watching sport be called a waste of time. To my mind, observing a sports game is a fascinating pastime. However, in my opinion they are mistaken. Last argument: Even some hidden talents can be discovered. Target sentence: Although, it is obvious that we should not spend much time in front of the TV, some people believe that watching sports can make a person obese.'</li><li>'Thesis: Some people think that spending free time watching sport on TV is just killing precious time. However, my personal opinion is that this activity is useful. Last argument: Also, it can be a good practice if you play some sport Target sentence: because while watching game you can learn some tricks of this game and then apply them in life.'</li></ul> |

## Evaluation

### Metrics
| Label   | Accuracy |
|:--------|:---------|
| **all** | 0.8741   |

## 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("Zlovoblachko/dim1_setfit_model")
# Run inference
preds = model("Thesis: In my opinion, watching sports on TV is a good opportunity to get relax. There are many reasons why some people like to watch sport games on TV. Last argument: None Target sentence: Where is the truth?")
```

<!--
### 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   | 13  | 62.7806 | 140 |

| Label | Training Sample Count |
|:------|:----------------------|
| L     | 540                   |
| H     | 540                   |

### 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: 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 |
|:------:|:----:|:-------------:|:---------------:|
| 0.0000 | 1    | 0.2674        | -               |
| 0.0014 | 50   | 0.3254        | -               |
| 0.0027 | 100  | 0.3135        | 0.3191          |
| 0.0041 | 150  | 0.2963        | -               |
| 0.0055 | 200  | 0.2799        | 0.2615          |
| 0.0068 | 250  | 0.2521        | -               |
| 0.0082 | 300  | 0.2554        | 0.2495          |
| 0.0096 | 350  | 0.2518        | -               |
| 0.0110 | 400  | 0.2484        | 0.2488          |
| 0.0123 | 450  | 0.2498        | -               |
| 0.0137 | 500  | 0.2456        | 0.2465          |
| 0.0151 | 550  | 0.2449        | -               |
| 0.0164 | 600  | 0.2433        | 0.2435          |
| 0.0178 | 650  | 0.2416        | -               |
| 0.0192 | 700  | 0.2424        | 0.2410          |
| 0.0205 | 750  | 0.2381        | -               |
| 0.0219 | 800  | 0.2302        | 0.2300          |
| 0.0233 | 850  | 0.227         | -               |
| 0.0246 | 900  | 0.2222        | 0.2428          |
| 0.0260 | 950  | 0.2129        | -               |
| 0.0274 | 1000 | 0.2138        | 0.2144          |
| 0.0288 | 1050 | 0.2026        | -               |
| 0.0301 | 1100 | 0.1888        | 0.2009          |
| 0.0315 | 1150 | 0.1735        | -               |
| 0.0329 | 1200 | 0.1658        | 0.2017          |
| 0.0342 | 1250 | 0.1646        | -               |
| 0.0356 | 1300 | 0.1442        | 0.1889          |
| 0.0370 | 1350 | 0.1428        | -               |
| 0.0383 | 1400 | 0.1169        | 0.1804          |
| 0.0397 | 1450 | 0.1237        | -               |
| 0.0411 | 1500 | 0.0989        | 0.1838          |
| 0.0424 | 1550 | 0.106         | -               |
| 0.0438 | 1600 | 0.102         | 0.1703          |
| 0.0452 | 1650 | 0.0823        | -               |
| 0.0466 | 1700 | 0.0822        | 0.1786          |
| 0.0479 | 1750 | 0.081         | -               |
| 0.0493 | 1800 | 0.0674        | 0.1685          |
| 0.0507 | 1850 | 0.0593        | -               |
| 0.0520 | 1900 | 0.0659        | 0.1732          |
| 0.0534 | 1950 | 0.0546        | -               |
| 0.0548 | 2000 | 0.0508        | 0.1889          |
| 0.0561 | 2050 | 0.0447        | -               |
| 0.0575 | 2100 | 0.0462        | 0.1637          |
| 0.0589 | 2150 | 0.0348        | -               |
| 0.0602 | 2200 | 0.0256        | 0.2151          |
| 0.0616 | 2250 | 0.0273        | -               |
| 0.0630 | 2300 | 0.0183        | 0.2285          |
| 0.0644 | 2350 | 0.0194        | -               |
| 0.0657 | 2400 | 0.0245        | 0.2068          |

### Framework Versions
- Python: 3.11.13
- SetFit: 1.1.3
- Sentence Transformers: 4.1.0
- Transformers: 4.54.0
- PyTorch: 2.6.0+cu124
- Datasets: 4.0.0
- Tokenizers: 0.21.2

## 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.*
-->