Labse Ru Sts

🚀 Base Bert for Semantic Text Similarity (STS) on GPU

A high-quality BERT model for calculating sentence embeddings in Russian. This model is based on cointegrated/LaBSE-en-ru, with similar context size (512), embeddings (768), and performance.

📦 Installation

To use this model, you need to install the transformers and sentencepiece libraries. You can install them using the following command:

pip install transformers sentencepiece

💻 Usage Examples

Basic Usage

Using with the transformers library

# pip install transformers sentencepiece
import torch
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("sergeyzh/LaBSE-ru-sts")
model = AutoModel.from_pretrained("sergeyzh/LaBSE-ru-sts")
# model.cuda()  # uncomment it if you have a GPU

def embed_bert_cls(text, model, tokenizer):
    t = tokenizer(text, padding=True, truncation=True, return_tensors='pt')
    with torch.no_grad():
        model_output = model(**{k: v.to(model.device) for k, v in t.items()})
    embeddings = model_output.last_hidden_state[:, 0, :]
    embeddings = torch.nn.functional.normalize(embeddings)
    return embeddings[0].cpu().numpy()

print(embed_bert_cls('привет мир', model, tokenizer).shape)
# (768,)

Using with the sentence_transformers library

from sentence_transformers import SentenceTransformer, util

model = SentenceTransformer('sergeyzh/LaBSE-ru-sts')

sentences = ["привет мир", "hello world", "здравствуй вселенная"]
embeddings = model.encode(sentences)
print(util.dot_score(embeddings, embeddings))

📊 Metrics

Evaluation on the encodechka benchmark

Tasks:

• Semantic text similarity (STS);
• Paraphrase identification (PI);
• Natural language inference (NLI);
• Sentiment analysis (SA);
• Toxicity identification (TI).

Performance and Sizes

Evaluation on the encodechka benchmark

Evaluation on the ruMTEB benchmark

📄 License

This project is licensed under the MIT license.