paraphrase albert small v2

Introduction

The paraphrase-albert-small-v2 model is part of the Sentence-Transformers library, designed for mapping sentences and paragraphs into a 768-dimensional dense vector space. This model is suitable for tasks such as clustering and semantic search.

Architecture

The model architecture includes:

• A Transformer component utilizing AlbertModel with a maximum sequence length of 100.
• A Pooling layer that applies mean pooling to word embeddings, which results in a 768-dimensional sentence vector.

Training

The model was trained by the Sentence-Transformers team, as mentioned in their publication "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks".

Guide: Running Locally

To use the model, follow these steps:

1. Install Sentence-Transformers:

   pip install -U sentence-transformers
   

2. Use the model with Sentence-Transformers:

   from sentence_transformers import SentenceTransformer
   sentences = ["This is an example sentence", "Each sentence is converted"]
   
   model = SentenceTransformer('sentence-transformers/paraphrase-albert-small-v2')
   embeddings = model.encode(sentences)
   print(embeddings)
   

3. Use the model 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/paraphrase-albert-small-v2')
   model = AutoModel.from_pretrained('sentence-transformers/paraphrase-albert-small-v2')
   
   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)
   

For enhanced performance, consider utilizing cloud GPUs such as those offered by AWS, Google Cloud, or Azure.

License

The model is licensed under the Apache-2.0 license, allowing for both personal and commercial use.