stsb xlm r multilingual

Introduction

The sentence-transformers/stsb-xlm-r-multilingual model is designed to map sentences and paragraphs into a 768-dimensional dense vector space, useful for clustering and semantic search tasks. It is part of the Sentence Transformers library, which enhances sentence similarity tasks.

Architecture

The model architecture consists of:

• A Transformer model: XLMRobertaModel with a sequence length of up to 128 tokens, maintaining the case of the input text.
• A Pooling layer that performs mean pooling over token embeddings to generate fixed-size sentence embeddings.

Training

This model was trained using a multilingual dataset, following the Sentence-BERT framework. It uses a Siamese BERT-network approach to derive sentence embeddings. Further information can be found in the related publication: Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks.

Guide: Running Locally

To run the model locally, follow these steps:

1. Install Dependencies:

   pip install -U sentence-transformers
   

2. Usage with Sentence Transformers:

   from sentence_transformers import SentenceTransformer
   
   sentences = ["This is an example sentence", "Each sentence is converted"]
   model = SentenceTransformer('sentence-transformers/stsb-xlm-r-multilingual')
   embeddings = model.encode(sentences)
   print(embeddings)
   

3. Alternative Usage with Hugging Face Transformers:

   from transformers import AutoTokenizer, AutoModel
   import torch
   
   def mean_pooling(model_output, attention_mask):
       token_embeddings = model_output[0]
       input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
       return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
   
   sentences = ['This is an example sentence', 'Each sentence is converted']
   tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/stsb-xlm-r-multilingual')
   model = AutoModel.from_pretrained('sentence-transformers/stsb-xlm-r-multilingual')
   encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
   
   with torch.no_grad():
       model_output = model(**encoded_input)
   
   sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
   print("Sentence embeddings:")
   print(sentence_embeddings)
   

4. Cloud GPUs Suggestion: For enhanced performance, especially with large datasets, consider using cloud-based GPUs such as those offered by AWS, GCP, or Azure.

License

This model is licensed under the Apache 2.0 License.