sentence-BERTino-v2-mmarco-4m

This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. It is a finetuned sentence-BERTino-v2-pt on ~4m mmarco examples.

Use query: and passage: as prefix identifiers for questions and documents respectively.

• loss: MultipleNegativesRankingLoss
• infrastructure: A100 80GB

If you find this project useful, consider supporting its development:

Usage (Sentence-Transformers)

Using this model becomes easy when you have sentence-transformers installed:

pip install -U sentence-transformers

Then you can use the model like this:

from sentence_transformers import SentenceTransformer

sentences = [
  "query: Questo è un esempio di frase",
  "passage: Questo è un ulteriore esempio"
]

model = SentenceTransformer('efederici/sentence-BERTino-v2-mmarco-4m')
embeddings = model.encode(sentences)
print(embeddings)

Usage (HuggingFace Transformers)

Without sentence-transformers, you can use the model like this:

1. pass your input through the transformer model
2. apply the right pooling-operation on-top of the contextualized word embeddings

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 we want sentence embeddings for
sentences = [
  "query: Questo è un esempio di frase",
  "passage: Questo è un ulteriore esempio"
]

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('efederici/sentence-BERTino-v2-mmarco-4m')
model = AutoModel.from_pretrained('efederici/sentence-BERTino-v2-mmarco-4m')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)

# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)