metadata

pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - feature-extraction
  - sentence-similarity
  - transformers
license: cc-by-4.0
language: bn
widget:
  - source_sentence: লোকটি কুড়াল দিয়ে একটি গাছ কেটে ফেলল
    sentences:
      - একজন লোক কুড়াল দিয়ে একটি গাছের নিচে চপ করে
      - একজন লোক গিটার বাজছে
      - একজন মহিলা ঘোড়ায় চড়ে
    example_title: Example 1
  - source_sentence: একটি গোলাপী সাইকেল একটি বিল্ডিংয়ের সামনে রয়েছে
    sentences:
      - কিছু ধ্বংসাবশেষের সামনে একটি সাইকেল
      - গোলাপী দুটি ছোট মেয়ে নাচছে
      - ভেড়া গাছের লাইনের সামনে মাঠে চারণ করছে
    example_title: Example 2
  - source_sentence: আলোর গতি সসীম হওয়ার গতি আমাদের মহাবিশ্বের অন্যতম মৌলিক
    sentences:
      - আলোর গতি কত?
      - আলোর গতি সসীম
      - আলো মহাবিশ্বের দ্রুততম জিনিস
    example_title: Example 3

BengaliSBERT-STS

This is a BengaliSBERT model (l3cube-pune/bengali-sentence-bert-nli) fine-tuned on the STS dataset.
Released as a part of project MahaNLP : https://github.com/l3cube-pune/MarathiNLP
A multilingual version of this model supporting major Indic languages and cross-lingual sentence similarity is shared here indic-sentence-similarity-sbert

More details on the dataset, models, and baseline results can be found in our [paper] (https://arxiv.org/abs/2211.11187)

@article{joshi2022l3cubemahasbert,
  title={L3Cube-MahaSBERT and HindSBERT: Sentence BERT Models and Benchmarking BERT Sentence Representations for Hindi and Marathi},
  author={Joshi, Ananya and Kajale, Aditi and Gadre, Janhavi and Deode, Samruddhi and Joshi, Raviraj},
  journal={arXiv preprint arXiv:2211.11187},
  year={2022}
}

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 = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)

Usage (HuggingFace Transformers)

Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.

from transformers import AutoTokenizer, AutoModel
import torch


#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    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 = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')

# 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)