---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
license: cc-by-4.0
datasets:
- nuvocare/WikiMedical_sentence_similarity
language:
- en
---

# WikiMedical_sent_bert

This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search.

WikiMedical_sent_bert is based on the 'all-MiniLM-L6-v2' sentence-transformers backbone and has been trained on the [WikiMedical_sentence_simialrity](https://huggingface.co/datasets/nuvocare/WikiMedical_sentence_similarity) dataset.

The model is able to predict whether two texts are related to the same wikipedia page, with only medical topic. 

## Usage (Sentence-Transformers)

Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:

```
pip install -U sentence-transformers
```

Then you can use the model like this:

```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('WikiMedical_sent_bert')
embeddings = model.encode(sentences)
print(embeddings)
```



## Evaluation Results

The model is evaluated on the test set of [WikiMedical_sentence_similarity](https://huggingface.co/datasets/nuvocare/WikiMedical_sentence_similarity). 
It achieves a :
- cosine spearman score of 0.86
- cosine pearson score of 0.94

For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=WikiMedical_sent_bert)


## Training
The model was trained with the parameters:

**DataLoader**:

`torch.utils.data.dataloader.DataLoader` of length 3170 with parameters:
```
{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```

**Loss**:

`sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss` 

Parameters of the fit()-Method:
```
{
    "epochs": 2,
    "evaluation_steps": 0,
    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 300,
    "weight_decay": 0.01
}
```


## Full Model Architecture
```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
  (2): Normalize()
)
```

## Citing & Authors

<!--- Describe where people can find more information -->