---
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity

---

# ellamind/e5small_sgb_aktg_bmf_experimental

e5 Small trained on the SGB5, AktG + train set of the BMF-Rundschreiben (19k queries in total).

For the finetuning Code, see [here](https://github.com/ellamind/gpt4gov/blob/main/gpt4gov/llama_index_finetune_embeddings.py).

For detailed analysis of top-line results see [this notebook](https://github.com/ellamind/gpt4gov/blob/main/notebooks/finetuning_results.ipynb).

This mdoel is the `e5small_tuned3_val_bmf` version.
![image/png](https://cdn-uploads.huggingface.co/production/uploads/647ef81ce9c81260ff84fdd7/8hVkoYUQuIB4ZFjC_dnz5.png)

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.

## Usage (Llama-Index)

```
embed_model = HuggingFaceEmbedding(
    model_name="ellamind/e5small_sgb_aktg_bmf_experimental",
    query_instruction="query: ",
    text_instruction="passage: ",
    device="mps",
    embed_batch_size=4,
)
```

## 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('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)
```



## Evaluation Results

<!--- Describe how your model was evaluated -->

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


## Training
The model was trained with the parameters:

**DataLoader**:

`torch.utils.data.dataloader.DataLoader` of length 258 with parameters:
```
{'batch_size': 32, 'sampler': 'torch.utils.data.sampler.SequentialSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```

**Loss**:

`sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss` with parameters:
  ```
  {'scale': 20.0, 'similarity_fct': 'cos_sim'}
  ```

Parameters of the fit()-Method:
```
{
    "epochs": 4,
    "evaluation_steps": 200,
    "evaluator": "sentence_transformers.evaluation.InformationRetrievalEvaluator.InformationRetrievalEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 103,
    "weight_decay": 0.01
}
```


## Full Model Architecture
```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, '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, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)
```

## Citing & Authors

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