Sentence Similarity
sentence-transformers
PyTorch
TensorFlow
ONNX
Safetensors
OpenVINO
Transformers
xlm-roberta
feature-extraction
text-embeddings-inference
Instructions to use sentence-transformers/stsb-xlm-r-multilingual with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use sentence-transformers/stsb-xlm-r-multilingual with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("sentence-transformers/stsb-xlm-r-multilingual") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Transformers
How to use sentence-transformers/stsb-xlm-r-multilingual with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/stsb-xlm-r-multilingual") model = AutoModel.from_pretrained("sentence-transformers/stsb-xlm-r-multilingual") - Inference
- Notebooks
- Google Colab
- Kaggle
Smaller Dimension needed 128
#4
by Koat - opened
This is a great model for cross language embedding but I would really like a smaller 128 or 256 dimension version that still has all the cross language support.
Hello, there are plenty of different embedding models in the official website of Sentence Transformers (https://sbert.net/docs/sentence_transformer/pretrained_models.html) where you can filter the list and find the most suitable model for you. In any case there is this one : "paraphrase-multilingual-MiniLM-L12-v2" Also available in HF, it is multilingual and produces 384 dimensional vectors.