---
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- SbertDistil
license: apache-2.0
datasets:
- wikimedia/wikipedia
- SiberiaSoft/SiberianPersonaChat-2
language:
- ru
- en
metrics:
- mse
library_name: transformers
---

# FractalGPT/SbertDistil

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.
This is a fast and small model for solving the problem of determining the proximity between sentences, in the future we will reduce and speed it up. [Project](https://github.com/FractalGPT/ModelEmbedderDistilation)

<!--- Describe your model here -->

## 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
import numpy as np
from sentence_transformers import SentenceTransformer
```

```python
model = SentenceTransformer('FractalGPT/SbertDistil')

def cos(x, y):
  return np.dot(x, y)/(np.linalg.norm(x)*np.linalg.norm(y))
```

```python
text_1 = "Кто такой большой кот?"
text_2 = "Who is kitty?"
a = model.encode(text_1)
b = model.encode(text_2)
cos(a, b)
```

```
>>> 0.8072159157330788
```

## Training

* The original weights was taken from [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2).
* Training was conducted in two stages:
1. In the first stage, the model was trained on Wikipedia texts (4 million texts) for three epochs.
   <img src="https://github.com/FractalGPT/ModelEmbedderDistilation/blob/main/DistilSBERT/Train/1_st_en.JPG?raw=true" width=700 />
3. In the second stage, training was conducted on Wikipedia, a dialog dataset, and NLI for one epoch.
   <img src="https://github.com/FractalGPT/ModelEmbedderDistilation/blob/main/DistilSBERT/Train/2_st_en.JPG?raw=true" width=700 />

## Full Model Architecture
```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 312, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
  (2): Dense({'in_features': 312, 'out_features': 384, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)
```