Abstract Sim Sentence Pubmed

🚀 Abstract Sentence Mapping Model

This model is designed to map abstract sentence descriptions to sentences that match those descriptions. It is trained on Pubmed sentences. You can use the load_finetuned_model function to load the query and sentence encoders, and the encode_batch() function to encode a sentence with the model.

🚀 Quick Start

To start using this model, you can follow the steps below. First, load the fine - tuned model and then use it to encode sentences.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModel
import torch

def load_finetuned_model():


        sentence_encoder = AutoModel.from_pretrained("biu-nlp/abstract-sim-sentence-pubmed", revision="71f4539120e29024adc618173a1ed5fd230ac249")
        query_encoder = AutoModel.from_pretrained("biu-nlp/abstract-sim-query-pubmed", revision="8d34676d80a39bcbc5a1d2eec13e6f8078496215")
        tokenizer = AutoTokenizer.from_pretrained("biu-nlp/abstract-sim-sentence-pubmed")
        return tokenizer, query_encoder, sentence_encoder


def encode_batch(model, tokenizer, sentences, device):
    input_ids = tokenizer(sentences, padding=True, max_length=128, truncation=True, return_tensors="pt",
                          add_special_tokens=True).to(device)
    features = model(**input_ids)[0]

    features =  torch.sum(features[:,:,:] * input_ids["attention_mask"][:,:].unsqueeze(-1), dim=1) / torch.clamp(torch.sum(input_ids["attention_mask"][:,:], dim=1, keepdims=True), min=1e-9)

    return features

Advanced Usage

The basic usage already covers the main functionality of the model. For more complex scenarios, you can modify the parameters in the encode_batch function, such as changing the max_length according to your specific needs.

| Property | Details |
|----------|---------|
| Tags | feature - extraction, pubmed, sentence - similarity |
| Datasets | biu - nlp/abstract - sim - pubmed |